IoT-based Obstacle Recognition Technique

for Blind Persons

Object detection is a critical task in computer vision, essential for real-time applications ranging from
autonomous vehicles to surveillance systems. This proposed work presents a comparative evaluation of
Single Shot Multibox Detector (SSD algorithms). YOLOv3, MobileNetv3, RetinaNet and Faster R-CNN, in
the context of real-time obstacle detection from camera images. The study evaluates these algorithms on
performance metrics Precision, Recall and F1 score across various Intersection Over Union (IoU) thresholds.
Also, the computational efficiency in terms of the time taken per frame is assessed to determine the
effectiveness of each algorithm. The workflow includes image processing, augmentation, and application of
SSD models to detect objects like vehicles, pedestrians and traffic signals. Results indicate that YOLOv3
achieves the highest precision of 97% demonstrating robust performance in real-time scenarios, while
MobileNetv3 follows closely with 94%, RetinaNet and Faster RCNN achieves accuracy 90% and 90%
respectively. These findings contribute to understanding the trade-offs between accuracy and computational
efficiency in selected suitable SSD models for practical deployment.

I. INTRODUCTION
Visual aids are the fundamental needs of the visually impaired for interaction with the environment, enabling them to navigate
their surroundings, recognize objects, and produce audible outputs [1] . The world needs to work on the unique challenges that
necessitate solutions. Object detection permits machines to recognize and identify the object in an image [2]. To solve this
problem, an obstacle detection and recognition system in a real-time environment is required. The affected people need this
type of device that can help them walk on the road to avoid major accidents. The techniques of artificial intelligence helped
them recognize objects and navigate the blind person both indoors and outdoors [3]. Advancements in deep learning techniques
have significantly improved the accuracy and efficiency of object detection algorithms. One such innovative approach is the
single-shot detector (SSD) algorithms [4] and two-shot detectors. Detection algorithms are used in a real-time object detection
framework that combines the strengths of both speed and accuracy. Traditional object detection methods involved multi-step
processes, including region proposal generation and object classification. However, single-shot detector (SSD) revolutionized
this by performing both tasks in one step, making it much faster. The features used in this proposed work are boundary box,
confidence score, and thresholding [5]. Studies have been conducted to detect the obstacles using sensors or cameras with
some machine learning and deep learning technologies. Mini Cameras used to detect obstacles from images in live systems
using machine learning techniques [6]. the Yolo algorithm using cameras and image dataset obstacles has been detected and
recognized to help the visually impaired avoid accidents and improve their walking in both indoor and outdoor environments.
The above techniques are used for detecting objects and text-to-speech for audio voice [7].
The motivation behind this proposed work is:
 In certain scenarios, the obstacle detection system may need to interact with humans or provide interpretable outputs.
Challenges include designing user-friendly interfaces and ensuring the system's decisions are understandable and
trustworthy.
 Over time, the reliability of an obstacle detection system may be affected by factors such as model degradation, changing
environmental conditions, and wear and tear of sensors. Ensuring long-term reliability and ease of maintenance is an
ongoing challenge.
 Achieving real-time performance in obstacle detection systems is crucial for applications like autonomous vehicles,
where timely responses are essential. Ensuring that the detection models meet real-time constraints without sacrificing
accuracy is challenging.

Addressing these challenges requires a multidisciplinary approach, combining knowledge from computer vision, machine
learning, and domain-specific knowledge for the application at hand [8]. Continuous efforts are essential to overcome these
challenges and advance the effectiveness of obstacle detection and warning systems using SSD models [9]. In the navigation
system, ultrasonic sensors installed in both shoes and a walking stick are used and algorithms are used to identify obstacles
and immediately inform the user of the presence of obstacles [10]. Most of the previous work in obstacle detection rates below
96%. This goes a long way to explain the rationale behind our proposed model, which seeks to solve the problems of the
visually impaired. Previous systems used computer vision algorithms to improve the accuracy of outdoor navigation systems
for the detection of obstacles during walking. However, these algorithms have been found to be incapable of providing the
desired results [11]. The objective of this work is to provide a comprehensive analysis of the performance of several advanced
SSD algorithms. YOLOv3, MobileNetv3, RetinaNet and Faster RCNN in the context of the obstacle detection and recognition.
Evaluation measures including precision, recall and F1 score will be compared on the various datasets and experimental
settings. By elucidating the comparative performance of these algorithms, this This study begins with the idea of object
detection typically for the visually impaired. This research focuses on the effectiveness of the single-shot and two-shot
detection algorithms. These SSF algorithms are considered for their capability, as a means of providing environmental
awareness.
This paper has the following contributions:

 The central aim of our study is to focus on the performance, strengths, and limitations of detection algorithms when
applied in a real-time environment.
 The potential impact of this study is to revolutionize assistive technologies, to make them more accessible and
effective concerning accuracy and time for the visually impaired.
 Through an in-depth analysis of prior works utilizing deep learning detection algorithms, we identified key
innovations and common defects. This critical examination of the strengths and weaknesses of various approaches,
informs efficient models.
 By providing a comprehensive evaluation and comparison of various detection algorithms, this study contributes to
the broader field of autonomous systems. These findings and recommendations presented in this paper have
significant implications for optimizing obstacle detection systems particularly in the context of mining operations.

The visually impaired faces challenges in their daily life, from navigation systems to avoiding the obstacle. Object
detection when combined with accessible interfaces, can allow them to cover these objects. Detection algorithms have gained
prominence for their speed and efficiency in real-time environments. Experiments in this paper show the comparative analysis.

The paper has been organized as follows.

 Section 1: The Literature review provides extensive details about the previous work on obstacle detection models.
 Section 2: The framework and architectures of detection algorithms are discussed in this section of the paper.
 Section 3: Implementation and Results of different detection algorithms from different environments.
 Section 4: Conclusion

II. Related Work

Bling people face many challenges in their daily routine, because of the dependency in their daily tasks. These challenges have
an impact on their independence, mobility, and overall quality of life. Some of the key challenges include: Safety is the biggest
challenge they face in their daily lives, without visual assistance they face the risk of accidents, falls, and collisions with objects
or other individuals. Navigation is one of the most prominent challenges for visually impaired individuals. They face
difficulties in road crossing, avoiding obstacles, Locating destinations, and Independence. The review encompasses various
technologies, methodologies, and man-made approaches utilized in developing these systems. It also highlights common
challenges and limitations while emphasizing the real-world impact of such technology [12].
 TABLE 1
Summary of Literature Review: Methods, Goals, techniques and Achievements
Refences Research problem Goal Techniques and Achievements
hardware’s
[13]-2024 To avoid collision WBSS is F-KAZE algorithm Prove good accuracy of the
designed for 7 m- proposed technology.
class USV.
[14]-2024 To navigate visually To enhance vision Ultrasonic sensor and Navigate perfectly for
impaired. and navigate RGC sensing camera. visually impaired.
blind people
[15]-2023 To remotely locate To achieve high GPS receiver, GSM Detection and
stick. security Modem, statistical automatically gives alter.
approach.
[16]-2023 To get directions by Object tracking Deep Neural Outperform.
hand Networks (DNN) and
YOLO version 5
[17]-2023 Alert for terrorist attack Sense the distant HC-05, ultrasonic Robot moves in all
in military. object and sensor, Arduino Uno directions.
monitor human microcontroller
voice.
[18]-2023 Selfy -mobility Movement with preprogrammed Health monitor system and
hand gestures. algorithm Alert system.

[19]-2023 To improve quality of To navigate and k-nearest neighbor satisfactory accuracy

life position. (KNN) algorithm
[20]-2023 Detect object timely. Monitor lifestyle. Statistical technique Improved accuracy.

[21]-2022 Easy life for common Wearable, TensorFlow detection Highly efficient and
individuals. modest, versatile algorithm convenient system.
system

As shown in Table 1, the existing literature covers a wide range of techniques, hardware configurations and goals aimed at
addressing techniques used in existing work and what they achieved. These approaches have achieved varying degrees of
success in different contexts. However, many of these methods rely on traditional techniques, while they offer notable
achievements, there are still gaps in terms of efficiency, accuracy and adaptability. Table 1 provides a comprehensive overview
of the boarder landscape of research efforts in this field. Details of the existing work is discussed below.
Jin, J., et al. [13] works on obstacle detection algorithm utilizing deep learning is proposed for an unmanned surface vehicle
(USV) equipped with a wide baseline stereovision system (WBSS)
Patel, I., M. Kulkarni, and N. Mehendale [14] have carried out various researches in a bid to develop gadgets that will be of
help to the visually impaired. This manuscript is devoted to different sensors applied to the navigation systems designed for
this category of people. The review looks at the current developments in the navigation aids for the visually impaired such as
the visual, proximity and LIDAR sensors among others. These sensors produce a lot of data and these data are then used to
mimic the environment. The manuscript also addresses the strengths of each type of sensor and the best prior that provide the
best outcomes. Some of the issues that are related to the use of sensors are also presented as well as some of the ways through
which they can be addressed.
module Rajesh, P., et al [15] used sensors to collect quantitative datasets to train the model. Blind is monitored by using a
mobile application for security and navigation. The system integrated a soil sensor, stair detection sensor, and obstacle sensor.
Radiofrequency is used to decide the obstacle-free track. Global positioning system (GPS) is used for navigation. Keywords—
Arduino Uno, Ultrasonic sensor, Infrared sensor, Soil moisture sensor, Global Positioning System (GPS), Global System for
Mobile Communication (GSM).
Jayachitra, J., et al [16] aims to develop a smart glove to give direction with hands. Deep neural networks and You Look Only
Once (YOLO) version 5 is used to train the model. It also includes cameras and microphones. The image-based dataset is used
in this paper to train the model. The contribution of this work includes finding the color of the object.
Sissodia, R., M.S. Rauthan, and V. Barthwal [17] aimed to help the visually impaired by using smart cars. The car is composed
of different sensors that detect objects and take directions in the form of a voice from the user. It also monitors the distance of
the user from the object. An application is developed to select the path and observe the movement of the car. The quantitative
data is collected by the sensors to measure the distance from the object. The application used Artificial intelligence to train the
model. This article is published in IEEE in the year 2023. Keywords— Arduino Uno, L298N, Bluetooth (HC-05), Ultrasonic
sensor (HC-SR04).
Chauhan, R., J. Upadhyay, and C. Bhatt [18] aimed to develop an IoT-based wheelchair for disabled persons. The machine
learning algorithm is used to make decisions about pathfinding more securely and safely. Sensors used in this paper collect
quantitative datasets for results
Lima, R., et al. [19] research significantly impacts the quality of life for visually impaired individuals by enhancing their ability
to navigate safely. Several additional aspects can be analyzed to better understand environmental directions and obstacles. The
feature matching process is supported by the brute force approach from K-Nearest Neighbor (KNN) algorithm. This approach
is essential in preventing hazardous situations
Prathibha, S., et al. [20] used Yolo algorithm in this work to increase the efficiency of the work. The Yolo family belongs to
the Convolutional Neural Network (CNN) algorithm. The work supports many problems the visually impaired face in their
surroundings. The white cane is used as an obstacle-detection device. Sensors are integrated with this white cane to take data.
The cane is connected with the mobile applications where the Yolo algorithm is trained. it increases the accuracy of the model
Masud, U., et al. [21] used Camera and detection sensors to detect the object. Raspberry 4B, camera, ultrasonic sensor, and
Arduino are integrated on a stick. Stick will be used by the user to detect the object; the dataset is transferred to the mobile
application for classification. The deep learning algorithm is used in this paper for classifications. TensorFlow Object Detection
algorithm is used for classification and to train the model. The ultrasonic sensor is also used to detect and send the beep to the
user, with the help of a camera they can send the details about the object. The device is designed to make the user more secure
in the crowd.

The challenges faced by the visually impaired limit their daily activities in interaction with the surrounding world. The
navigation system is the biggest need for the visually impaired to give them directions about their path. The IoT field plays a
vital role in helping the blinds in their walking in outdoor and indoor environments. Dhou, S., et al. [22] designed a mobile
application to process the data. Sensors collect their data and send it to the mobile application for preprocessing and
classification results. Google Maps is used to give them directions. Its prototype is developed and tested in different
environments. Accuracy is improved from 81% to 99%. CNN model is compared with the KNN classifier to improve its
performance. KEYWORDS: Visually impaired, walking assistance, IoT, machine learning.

Table 2. presents a discussion on both innovations and defects. It outlines the technologies employed in the referenced research
articles, along with their respective weaknesses. This comparative analysis aims to see the novelty of the proposed work.
Research questions for the comparative study on obstacle detection for the visually impaired are:

RQ1: How do different detector algorithms perform in terms of accuracy and reliability for obstacle detection across various
environmental conditions?
RQ2: How does the performance of deep-learning-based detector algorithms compare in the obstacle detection model, and are
there scenarios where one approach outperforms the other?

Building on this foundation, more recent advancements have focused on improving object detection through specialized
methods, particularly one-shot and two-shot detectors. These techniques have introduced innovations that seek to address some
of the limitations found in earlier works. Table 2 narrow the scope to analyze these specific methods, highlighting their
innovations and identifying defects that persist despite their novel contributions. By examining these newer approaches, we
can better understand how they improve upon or struggle with the challenges discussed in the boarder literature from Table 1.
 TABLE 2
Innovations and Defects in Existing Studies Based on Table

Type Object Detection Algorithms Innovations Defects

One shot detector YOLOv3 An algorithm is powerful and Emphasize a deeper focus on
efficient for target detection implementation comparisons,
[23]. including scenario analysis.
One shot detector Efficient-Net Several techniques used Neural architecture search
including weight decay, mosaic technology (NAS) to improve the
Decay, Mosaic Data universality and performance of
augmentation and loss detection.
normalization to improve the
performance of Efficient-Net
[24].
Two shot R-CNN It is more precise to state that it Training time is too long.
detectors amalgamated deep learning
methods and traditional
methods [25].
RetinaNet Retina-Net identifies objects of The Kalman Filter algorithm was
One shot detector different sizes by using a employed to enhance the
feature pyramid network and accuracy of object detection
focal loss [26]. results.
One shot detector MobileNetv3 UAV device based object Reduced model accuracy,
detection model [27]. suboptimal real-time
performance and lack of
sensitivity to image.
Two shot Center-Net By appropriately reducing Generate incorrect boundary
detectors structural complexity, Center- boxes.
Net attains a favorable [28].
Two shot Faster-RCNN Faster R-CNN employs an Gives low accuracy when used
detectors innovative approach by with VGG16 network.
utilizing the convolution
network to generate proposed
boxes. This strategy reduces the
number of frameworks from
2000 to 300 [29].
 Innovations and the defects found in the previous research work are discussed in Table 2. Models that have used one-shot
and two-shot detectors are cited in the above table with findings.

III. MATERIALS AND METHODS

In recent times, the deep learning-based object detection field has witnessed the emergence of numerous high-performance
methods. These approaches specifically consider the unique characteristics of road traffic [30], frequent changes in lighting
and background conditions [31] , diverse dimensions and positions of obstacles in images [32] , as well as the shape, size, and
color of the obstacles [33]. Given the demanding real-time performance requirements for object detection algorithms selected
from a multitude of options. The aim is to conduct a comparative analysis of these algorithms, emphasizing their exceptional
accuracy and real-time capabilities.
The preceding section elucidates the comprehensive design of the obstacle detection model, comprising two distinct segments:
the input phase and the analysis phase. The input phase involves the integration of a sensing device camera to observe the
surroundings, while the analysis phase unfolds in the python Jupiter environment. This section delves into the implementation
details of the Obstacle Detection model. The obtained data consists of values of frontal image. The complex dataset is sent to
the microcontroller to be fed into system. The real-time experiments have been performed using the dataset received from
outdoor and indoor. The model was successfully functioning during real-time experiment. The gathered dataset serves as the
foundation for generating results through Python. The dataset comprises information collected from camera monitoring the
surrounding. To enhance the accuracy of the results, the dataset undergoes preprocessing in the Jupiter environment, ensuring
a robust foundation for subsequent analysis and interpretation.

A. Architecture of the proposed model

The Framework of Object detection algorithms typically involves several key components and stages. Figure 1 shows the
external architecture of the proposed model.

3.Region 5. Post
Two shot detector
Proposal processing steps

1. Image capture 2. Preprocessing 6. Output 7. Evalutation 8. Fine-Tuning

One shot detector 4. Prediction

FIGURE . Internal Architecture of the proposed Obstacle Detection Model.

The overview of the external architecture for an obstacle detection system using the mini camera, Arduino Uno, Bluetooth
device, mobile application, and hands-free module is shown in Figure 1. The mini cameras capture the image of the
environment. Images are processed to detect the obstacles. This processing can be done directly on the camera, in this proposed
model we have fixed the model on knee gloves which makes it easy for the user to detect all the surroundings. The Arduino
Uno is the control unit for the proposed system. The algorithms are implemented on the Arduino Uno to make decisions based
on the input. These algorithms used techniques for object recognition and edge detection to identify the obstacle. The mobile
application is used as the user interface, it receives information from the Arduino Uno and displays information about the
detected obstacle to the user. The application also provides additional information about the navigation system.
Figure 2 illustrates the internal architecture of the proposed mode, details are discussed below:

 Image Capture: The process begins by capturing an image sent to the dedicated mobile application developed for this
purpose illustrated in Figure 2.
  Processing: Upon receiving the image, the server performs preprocessing steps to enhance the quality of the image and
prepare it for further analysis. This involves considering input size, anchor box configuration, and feature map
resolution. The preprocessed image is passed through a pre-trained classification model. The model's purpose is to
classify the content of the image, in this case, identifying specific objects and patterns in indoor and outdoor
environments.
 Region Proposal (Two Stage Detector): Two-stage detectors, a region proposal network (RPN) may be employed to
generate candidate bounding boxes or regions of interest (Roles) that are likely to contain objects.
 Prediction (One Stage Detector): For a one-stage detector, predictions for bounding boxes and class probabilities are
directly made in one step through the network, eliminating the need for a separate region proposal stage.
 Post-processing steps: post-processing steps may include non-maximum suppression (NMS) to filter redundant or
overlapping bounding boxes and improve the final set of detections.
 Output: The final output consists of the detected objects, along with their corresponding bounding boxes and class
labels.

Evaluation: the algorithm’s performance is evaluated by using different performance metrics such as precision, recall, F1 score,
and others. This is often done on a separate validation or test dataset

B. Data Preprocessing and Augmentation

Image processing is a process of detecting individual objects and images. Sometimes in image classification cannot get better
results because the image can have various noises or the image has many other objects that are not of interest [34]. The input
image is used in Python to do preprocessing and augmentation it converts the image into the mask. In the case of an outdoor
environment, there can be multiple things in the image which will give the mask to identify the multiple objects [35]. This
approach is mostly used in self-driving cars [36], because it's not only important to detect the objects in front of you it's always
important to detect the shape of the object in front of you to avoid the collision. So, this technique will help the visually
impaired in their walk on the roads. This method helps in classifying the important objects. Figure 3, presents the extraction
of meaningful images reduces noise, and improves the accuracy of the model. Image classification algorithms focus on the
highlighted objects and help in recognizing them by assigning the class labels [37].
After the preprocessing of the image augmentation is used on training dataset [38], which helps image to improve robustness
and generalization ability of machine learning models. The process involves applying various transformations to the original
images [39]. It rotates the image into different degrees shifting the image along x and y axis, resize the image applied a shearing
transformation, which skews the image. Flipping the image horizontally and vertically. Increased and decreased the brightness
modified the contrast of the image. Change the intensity of color. Shifted the colors of the image. Gaussian noise distribution
is used cropping the random portion of the image. Randomly hide the parts of the image to simulate partial occlusion. Applied
motion blur and change the perspective of the image as if viewed from a different angle. Adjusted the gamma value to control
the overall brightness of the image. All this work of preprocessing and augmentation is done in python Jupiter. This process is
applied on 100 images as training dataset.

IV. RESULTS
The model is trained and tested on real-time objects. It measures the discontinuity in the image by measuring point edges and
lines. It measures the high frequency of objects and ignores the noise in it [40]. Figure 3 shows the results from preprocessing
and augmentation. Images have been resized to fix in SSD model expects. Scale the pixel values to a range that the model can
handle, usually between o and 1. This is done by dividing the pixel values by 255. Subtracted the mean RGB values calculated
from the training dataset to center the data. This helps in improving the model’s convergence during training. Applied data
augmentation techniques random cropping, flipping, rotation and color jittering to make the model more robust to variations
in the input data.
.

A. EXPERIMENTAL RESULTS OF PREPROCESSING ANG AUGMENTATION

The experimental results presented in this section were derived using a python environment, specifically Jupyter Notebook,
which provides a user-friendly interface for developing and testing out model. The libraries utilized in this process include
NumPy for efficient numerical computations and TensorFlow for building and training the deep learning model. This setup
allowed for a streamlined workflow in managing the complex operations involved in preprocessing and augmenting the dataset.
The visual data is captured by a camera integrated into specialized glasses designed for visually impaired users. This data is
transmitted to an Arduino Uno via Bluetooth, allowing real-time processing. The preprocessing phase is crucial for ensuring
that the input data is optimized for the model. All images captured by the camera are resized to a uniform dimension,
specifically 224*224 pixels, which is the expected input size for our model. This resizing not only enhances processing speed
but also maintains consistency across the dataset.
Data augmentation is employed to enhance the diversity of the training dataset, which plays a vital role in improving the
model’s generalization capabilities. Rotation, Flipping, Zooming and brightness adjustment is applied to the images.

Figure 3 illustrates the step-by-step image transformation process in the system, showcasing how raw outdoor environment
images (a), (b), (c), (d) are preprocessed and augmented to enhance the performance of the object detection model. The 1st
column of the table is the raw, unprocessed image of the outdoor environment, captured directly from the sensor/camera. The
effectiveness of the preprocessing and augmentation strategies was evaluated by training the model on the augmented dataset.

B. IMAGE CLASSIFICATION BY USING SSD MODELS

The process of detecting the right object at the right time is the most important part of this model. Any negligence in measuring
the parameters can cause an accident or any other loss [41]. This intelligent real-time obstacle detection model is proposed to
avoid accidental situations. Its notification system is an intelligent approach to guide the visually impaired about their current
situation to avoid collision. When a color is fed into the input layer of one-stage and two-stage algorithms, the algorithm goes
through a series of steps to perform object detection. After image segmentation, some detection algorithms are used to compare
the performance concerning time and accuracy.
Steps used by detection algorithms to recognize the object from the image:
 VGG-16 Network: This network is used to extract the features from the original image.
 Feature Extraction: It is used on resized and preprocessed images, this image is passed through the one-stage model's
convolutional layers, which serve as a feature extractor. These layers capture hierarchical features at different scales in the
image. The network layer recognizes patterns, edges, textures, and object-specific features.
 Anchor box generation at various levels of the features hierarchy, the one-stage algorithms generate a set of anchor boxes
(default bounding box). These boundary boxes have different aspect ratios and scales to capture objects of varying sizes and
proportions in the image. Prediction, for each anchor box, the one-stage model makes predictions regarding two key aspects.
 Class Score: The model assigns a class score to each anchor box, indicating the likelihood that the box contains an object of
a particular class. These class scores are obtained through classification layers.
 Bounding box: The model predicts the bounding box offsets to adjust the anchor boxes and better fit the actual objects.
These offsets are obtained through regression layers. The one-stage algorithm can detect multiple objects in a single forward
pass through the neural network, making it highly efficient for real-time applications. It leverages anchor boxes to handle
objects of various sizes and aspect ratios and produces reliable object detection results.
 Final Predictions: the final predictions of the one-stage algorithm consist of the remaining bounding boxes, their
correspondence class label, and confidence score. These predictions indicate the objects detected in the input image and their
locations.
Performance metrics such as accuracy, precision and F1-score were calculated to assess the model’s performance. The results
indicates that the integration of preprocessing and augmentation techniques led to significant improvement in these
performance metrics compared to models training on non-augmented datasets. For instance, the model’s accuracy is
increased.
To validate the performance of the SSD model, a comprehensive evaluation was conducted using a test dataset comprising
images representative of various real-world scenarios. The dataset was divided into training, validation and test sets ensuring
that the model was trained on diverse examples while being evaluated on unseen data. A typical split ratio of 70% training,
and 15% testing was applied. Using pretrained models allows the advantage of leveraging prior knowledge gained from
 large datasets, which accelerates the learning process and improves the model’s detection ability. Especially in complex
real-time environments. The results from RetinaNet, mobileNetv3, YOLOv3 and Faster RCNN were compared and the
algorithms gives its best selected for the model training. YOLOv3 was particularly effective in detecting small and
overlapping objects, which is vital for obstacle detection in complex and cluttered environment.
By integrating YOLOv3 into the proposed system the model was able to deliver superior performance, providing timely and
accurate feedback to users. This ensures that obstacles are detected and communicated to the visually impaired individuals
without delay, enhancing both safety and navigation efficiency.

The one-stage and two-stage algorithms depend on many factors like model architecture, the hardware used in the model, and
the implementation techniques in Figure 4. Various scales are depicted in the results, indicating the confidence levels for object
detection. These scales represent the threshold 0.5 is applied to determine whether a detected objects is considered a valid
detection, highlighting the trade-offs between precision and recall. By adjusting these scales, users can optimize detection
performance based on their specific requirements and the context of their application.
Mathematical key equations and concepts involved in each algorithm:
RetinaNet is focused for its focal loss that is useful in dealing with the class imbalance in object detection.
Equations used by the model:
Focal Loss = − (1 − ) ( ) (1)
 Where Tp is the predicted probability for the true class, wt is the weighted factor for class C1 and f is the focusing parameter
to adjust the importance of hard examples.
( , ∗) = {0.5( − ∗)2 ∣ − ∗∣ −0.5 ∣ − ∗∣< 1 (2)
Where x is the predicted bounding box coordinates and x* is the ground truth coordinates.
MobileNetv3 is designed for efficient mobile vision applications with reduced computational complexity and increased
accuracy:
Depthwise Convolutional:
( )= ∗ , = 1…,! (3)
Where Bc is the c-th input channel, Lx is the corresponding filter, and * denotes the convolution operation.
Pointwise convolutional:
Outputp = " # $ % #& ('( ( )) = % #&1 ∗ 1 . '( ( ) (4)
Where Outputp is the output feature map after the pointwise convolution. PointwiseConv represents the pointwise convolution
operation. Outputd is the input feature map (usually the result of a Depthwise convolution in the MobileNet architecture).
Conv1*1 represent a convolution operation with kernel size of 1*1.
Yolov3 algorithm is a real-time object detection system that frames object detection as a single regression problem, straight
from image pixels to bounding box coordinates and class probabilities.
For each grid cell, YOLOv3 predicts:
" ( *$$ / ',- ) = * *$$ / . , = 1 * *$$,, (5)
Final score calculated for each class in a bounding box is the product of the objectness score and the class probability:
/( *$$ ) = " ( ,- ) ∗ " (% *$$ / ,- ) (6)
Yolov3 uses a multi-part loss function:
Loss = 0 ) .1 ,- . [1'3" + 5/6 ) ] + 0# ,- .1 # ,- . 5/6 (7)
Where IOUPt is the intersection Over Union between predicted and ground truth boxes, 5/6 ) is the mean squared error
for coordinates, CEclass is the Cross Entropy for classification and 0 terms are weights.
Faster RCNN introduces Region Proposal Networks (RPN) to generate region proposals and then applied Fast R-CNN for
object detection.
Class prediction probability:
( *$$*) = $ / ∑ - = 1 $ (8)
$$9*$ :%;; = /< ℎ 1 $$(! ( . >" )) (9)
Where smoothL1 aims to make the model robust to outliers. L1 Loss reduces sensitivity to outliers.
C. PERFORMANCE METRICS RESULTS
The performance of an algorithm also depends upon factors such as input image resolution, input size, and hardware
acceleration. To determine the fastest algorithm in this proposed work, an experiment is conducted on images and results have
been compared. Different factors have been considered like accuracy and the time of detection taken by each algorithm.
 Detection time comparison of SSD Detection Algorithms.

MobileNetv3 and YOLO v3 performs better than RetinaNet and Faster RCNN. That’s all during the experiment, detection
time was tracked with a bar chart. Figure 5 shows object detection times for different SSD algorithms when working on fixed
number of images or frames. It is important to note that this comparison is essential in determining how efficient and effective
each algorithm may be in real-time applications. The y-axis represents the time taken by each algorithm to process the same
number of 100 images with time measured in seconds. To make comparison fairly, all algorithms were tested on identical
images collected [42]. As per this graph, every bar represents detection time for a particular algorithm while the height of each
bar depicted how long it takes to process those pictures. The YOLOv3 algorithm has the shortest detection time which is 7
seconds. Thus, it can be concluded that YOLOv3 is quite good at processing images efficiently hence being suitable for real-
time applications which require high speeds. MobileNetv3 takes 8.7 seconds as its detection time which is just slower that
MobileNetv3 indicating that when it comes to speed RetineNet may not be efficient but there may be other advantages
associated with great accuracy and robustness. Faster RCNN algorithm has the longest detection time about 9 seconds
compared to others. This appears slowest amongst them. Among other things, this visualization shows how sure each of these
algorithms are with regards to its object detections and how these levels of belief vary. The model has been trained for 80
classes. Class names involves all possible objects a user might come across on the road while walking around. This
visualization helps to understand how confident each algorithm is in its object detections and the variability in confidence
levels. The x-axis represents the detection confidence score, which ranges from 0 to 1. These scores indicate the probability
that a particular class belongs to an object detected by a model while y-axis shows the frequency of such confidence scores
within dataset. Hence, this graph combines histograms and Kernel Density Estimation (KDE) plots for each algorithm that
provide smooth curves representing distributions of their confidence scores.
YOLOv3 Algorithm: This gives confidence store which is fairly high, with many detections having confidence scores
beyond 0.80, the KDE plot shows that there is a peak around 0.90, hence there is a strong confidence in its detections.
MobileNetv3: MobileNetv3 has a confidence score above .80 with a few peaks around 0.90 this indicates it reliable similar
to YOLOv3.
RetinaNet: RetinaNet exhibits a slightly wider spread in the values of confidence scores observed in RetinaNet due to a
definite peak at about 0.85. Although many detectors are assured, they have more variability unlike those found in YOLOv3
and MobileNetv3
Faster RCNN: Faster RCNN shows confidence scores with a peak around 0.87, indicating strong detection confidence.
Similar to the other algorithms, Faster RCNN demonstrates high confidence in its detection.
The high confidence score across all algorithms suggests robust performance in object detection tasks illustrated in Figure 6.
YOLOv3 and MobileNetv3 shows particularly high and consistent confidence level, while RetinaNet and Faster RCNN also
exhibit strong confidence with slightly more variability. YOLOv3 achieved 96% accuracy, MobileNetv3 achieved 92%,
RetinaNet and Faster RCNN achieved 90% accuracy on training data.

The model has been experimented with on the road and trained with the detection algorithm which gives the best performance
as compared to others. In the experiment on the road, dataset has been collected to check the accuracy of the trained model.
The above line chart in Figure 7 compares the performance of four different object detection algorithms, Yolov3, RetinaNet,
MobileNetv3, Faster RCNN across various performance metrics and threshold variations. The performance metrics evaluated
are Precision, Recall and F1 score. The threshold represents different intersection over union (IoU) values used to determine
true positive detections. The y-axis represents the scores of three performance metrics: Precision, Recall, and F1 score. Each
metric provides a different preceptive on the model’s detection accuracy and reliability. Measuring the accuracy of SSD
algorithms for obstacle detection involved the performance metrics that evaluate different aspects of the model’s performance.
Equations used by the model to measure accuracies are:

?@AB CDEFGFHB(?I)
" $ #= (9)
?@AB IDEFGFHB(?I)JKLMEB IDEFGFHB(KI)

It measured a single measure of the model’s accuracy across all object classes. Precision values for all classes. It provides a
single measure of the model’s accuracy across all object classes.
?@AB IDEFGFHB (?I)
: * = (10)
?@AB IDEFGFHB (?I)J KLMEB NBOLGFHB (KN)

Recall measured the proportion of true positive detections out of all actual obstacles (true positives and false negatives).
I@BPFEFDQ∗RBPLMM
91 $ =2∗ (11)
I@BPFEFDQJRBPLMM

The legends on the chart identifies each algorithm and the corresponding performance metric, helping to distinguish between
the different lines.

The Area Under the Curve (AUC) score is used to evaluate the performance of a classification models. Specifically, the AUC
score represents the area under the Receive Operating Characteristics (ROC) curve shows in figure 8, which plots the true
positive against the false positive rate at various Threshold settings. We have tested on different images.
(a)

(b)

(c)

(d)
 FIGURE 9. Detection error tradeoff graphs of classifiers.

The graph is figure 9 illustrates the false positive rate and false negative rate. It helps to understand the tradeoff between
detection errors and useful in evaluating and comparing the performance of different classification models.
YOLOv3: YOLOv3 shows strong precision values, but its recall and F1 scores exhibit variability as compared to precision.

MobileNetv3: The Precision values for MobileNetv3 are consistently high across all thresholds, indicating robust performance
in detecting and recognizing obstacles, while it is precision, it may miss some obstacles at certain threshold. Recall and F1
scores also remain stable, suggested good overall accuracy and balance between precision and recall.

RetinaNet: RetinaNet’s precision values also lower than those of MobileNetv3 and YOLO, but its precision shows an upward
trend, suggesting it becomes more effective at identifying obstacles as the IoU threshold increases. The F1 score reflect this
improvement in recall.

Faster RCNN: Faster RCNN’s Precision demonstrate high values similar to MobileNetv3, with recall and F1 scores that are
stable and slightly low, indicating a reliable performance across different threshold.

FIGURE 10. Detection Scores for various classes using different SSD algorithm.
Figure 10 shows a heatmaps that illustrates the detection scores for various object classes across ten different images. This
visualization provides an intuitive understanding of how well each object detection algorithm performs on different classes in
the dataset. The dataset consists of detection scores or frequencies for eight object classes: car, person, bicycle, truck, rikshaw,
footpath, traffic lights, bus [43]. These scores are recorded across ten images 1 to 10. Each cell in the heatmaps represents the
detection score for a particular class in a specific image. The x-axis of the heatmap lists the object classes and y-axis shows
the image numbers. Each cell’s intensity corresponds to the detection score, with darker shades indicating higher scores.
Annotating each cell with the actual detection score value helps to precisely understand the model’s performance [44]. The
heatmap allows for a quick comparison of detection scores across different classes. For instance, it can be seen if certain like
car, person consistently have higher detection scores across all images. The performance across the different images can be
evaluated. If some images have low detection scores for all classes, then it could be due to some conditions like low light or
occlusion in the image.
FIGURE 11. Confusion matrix based on classification performance of SSD models.

In this proposed work, confusion matrix has been computed to evaluate the classification performance of each object detection
algorithms. In each confusion matrix in figure 11 the horizontal axis represents the predictable label while the vertical axis
represents the true label of the dataset. The confusion matrix provides the performance metrics of the YOLOv3 in relation to
the ground truth label. The matrix shows the number of true positive, true negative, false positive and false negative of each
class [45]. The MobileNetv3 confusion matrix gives information about the classification of the algorithm. It focuses on the
aspects of the model that performs well or poorly in identifying objects within the dataset. The confusion matrix of the
RetinaNet algorithm measures the accuracy of the algorithm in different classes of objects. It points out the areas of
improvement in terms of the detection of specific classes and the areas of strength [46].
The confusion matrix of Faster RCNN measures the accuracy of the algorithm in classification. It helps in knowing the degree
of separation of the model in terms of object classes from the prediction it gives. These matrices provide a quantitative
comparison of each algorithm’s performance, which facilitates the identification of the best approach to object detection. The
detection model parameters include confidence score, class score threshold, input size image and input image scale.

TABLE 2
Comparison based on 2 Categories
Training Models Speed Accuracy(correct)
(How
Fast)
RetinaNet (SSD) 9.5ms 90%
MobileNetv3(SSD) 7.2ms 92%
YOLOv3 10.1ms 96%
Faster RCNN 11.8ms 90%

The above Table 2 illustrates the concluded results of all the detection algorithms that have tested after findings of different
results its concluded that YOLOv3 gives better results in proposed model, when trained in real-time environment application.
All the detection algorithms were good in different aspects but the model is trained with YOLOv3 which proves the best in all
aspects.

TABLE 3
Accuracy Comparison with existing work
Reference Accuracy Techniques (Existing work)
(Existing
work)
[47] Accurately Single Shot multibox
navigating detector
space
[48] Reliability YOLOR algorithm
and Safety
[49] 90.49% MobilNetv3 and SSD
[50] 0.781 MobileNet and SSD
[51] safety and YOLOv7
efficiency
Proposed 96% YOLOv3
work (our
study)
Table 3 illustrates that accuracies achieved in previous work and the accuracy of proposed model. YOLOv3 in proposed study
shows a slightly improved in Precision and F1score compared to existing work, indicating better overall detection accuracy.
The Recall is also marginally higher, suggesting a better balance between precision and recall. All evaluated algorithms in
proposed study show improvements over existing work, suggesting advancements in object detection techniques and algorithm
optimizations. Table 3 provides the comprehensive comparison of the accuracy of various SSD algorithms in proposed study
against existing work [52]. The improvement observed in proposed study stress the efficiency of the optimizations and
techniques that have been applied to these algorithms and can be useful for choosing the most suitable algorithm for a given
application. Further research should also extend to enhance these algorithms in order to make it more accurate and faster.
D. DISCUSSION
The information comparison of four SSD algorithms YOLOv3, MobileNetv3, RetinaNet and Faster RCNN was relevant to the
real-time object detection. The assessment focused on two critical aspects: Detection Time, along with the confidence score.
1) Detection time:
The study provides a comprehensive evaluation of SSD algorithms for obstacle detection and recognition, focusing on four
prominent models: Some of the instance augmentation models include: YOLOv3, MobileNetv3, RetineNet and Faster RCNN.
This analysis was performed with a dataset that contained several object classes and the evaluation of the algorithm was made
based on the model’s precision, recall and F1 score. The findings show that this article YOLOv3 produces the most accurate
results with the highest precision of 96%, recall and F1 score of 90%. MobileNetv3 second to list with an accuracy of 92%,
RetineNet and Faster RCNN with 90%. Thus, the primary outcomes of this investigation can be rendered as follows: YOLOv3
model outperform in concerns to accuracy and inference time making it optimal for real-time use. MobileNetv3 represents the
accuracy /inference time ratio lower than YOLOv3, RetinaNet and Faster RCNN are promising for the cases where the detector
should be both highly accurate and relatively fast.
2) Detection confidence:

The confidence score of YOLOv3 is high and very stable, with the maximum value of approximately 96%, this shows that the
equipment used in the detection of the disease is highly reliable. This high level of confidence minimizes false positive and
negatives, are critical in accurate obstacle recognition. MobileNetv3 shows confidence scores comparable to that of YOLOv3
with the score lower to around 92%. This suggests that YOLOv3 can be gives good detections, which makes it a good candidate
for applications where detection accuracy is important. RetinaNet showed a slightly wider spread in confidence scores, with a
peak around 0.90, while this indicates a bit more variability in detecting confidence, RetineNet still offers robust performance,
especially in scenarios where detection variability can be tolerated. The confidence score of Faster RCNN peaking 0.90 suggest
strong detection reliability. Despite its slower detection time, the high confidence scores justify its use in applications where
accuracy is prioritized over speed.
3) Limitations:

While this study provides valuable insights into the performance of various SSD models for real-time obstacle detection,
several validity threats must be considered.

Internal Validity:
The simulations were performed in a controlled environment, where the dataset was consistently processed and the evaluation
conditions were kept stable. However, internal validity might be threatened by potential biases introduced by the dataset itself.
For instance, inform preprocessing and fixed hyperparameters across models may favor one model over another, and such
biases could affect the comparative performance results.

Overfitting was mitigated using standard validation techniques, but no additional steps like cross-validation or dropout
regularization were applied. Therefore, there is a risk that the performance of certain models, especially the YOLOv3 could
be too optimized for the specific dataset used in this study, and might not generalize as well as to different datasets.

External Validity:
The results of the study were based on a specific hardware and software configuration, which means that the conclusion drawn
about the models’ inference times and processing capacities may not hold in different computational environments. Variations
in GPU performance, memory and network connectivity could significantly impact real-time performance, particularly when
deploying these models in the field.

Another threat to external validity in that the experiments were conducted on simulated data and under data conditions. Real-
world scenarios often involve dynamic environments, moving objects, and occlusions, all of which can influence detection
performances in ways that were not captured in the simulated setup. Therefore, the performance of YOLOv3 and other models
could differ in real-world applications, where conditions such as lighting, weather and motion might present additional
challenges.

4) Assumptions:

Few assumptions were made during this study:

 It was assumed that the object classes in the dataset were uniformly distributed. In real-world scenarios, certain object
types may be more prevalent than others, which could affect the detection performance.
 The study primarily considered static object in the evaluation process. Moving objects or objects with significant motion
blur may present additional challenges that were not fully addressed in this analysis.
 It was assumed that the preprocessing steps (such as resizing and normalizing) were consistently applied across all
algorithms. Variations in preprocessing techniques could influence the detection performance.
 The default hyperparameters of each SSD model were used in the evaluation. Fine-tunning these hyperparameters for
specific datasets or use cases could yield different results.

5) Recommendations:

Based on the findings and limitations of this study, several recommendations are proposed for future research.

Future work should involve a more diverse and comprehensive dataset that includes various object types, lighting conditions
and environmental settings to improve the generalizability of the results. Conduct extensive testing in real-world scenarios to
evaluate the robustness and reliability of the algorithms under different conditions, such as varying weather, occlusions and
dynamic backgrounds. Explore hardware-specific optimizations and model compression techniques like quantization and
pruning to improve the real-time performance of the models on different devices. Investigate the integration of advanced
techniques such as attention mechanisms, multi-scale feature integration and hybrid models to enhance detection performance
and efficiency.

V. CONCLUSION
The primary objective of this work was improving the performance of obstacle detection systems, by making them more
reliable, accurate and thus perform for real-time applications. To the best of our knowledge, this study offers a comprehensive
comparison with respect to recent algorithms Faster RCNN, MobileNetv3, RetinaNet and YOLOv3. We provided evaluation
according to theory accuracy and time efficiency which provide a comprehensive experience on how the models will be acting
in real-time applications. The practical approach ensures that findings based on such a method can be directly applied to real-
time situations. This includes autonomous navigation and safety systems. Otherwise, fields used to judge the model’s
effectiveness in dynamic, arduous circumstances. Our model employs efficient YOLOv3 detection algorithm to process images
and videos in real-time. These algorithms have been proven effective in our model. We acquired the real-world dataset that
include various real-world scenarios and obstacle types, this was difficult to train and test the model. MobileNetv3, despite
good balance between accuracy and speed, suitable for scenarios requiring both. Future work should address the limitation
identified and explore the recommendations provided to enhance the robustness and applicable of SSD-based object detection
models in diverse real-world scenarios.

VI. FUTURE WORK

While this study has provided valuable insights into the performance of various SSD algorithms for obstacle detection
and recognition, several avenues for future research can further enhance the applicability and robustness of these
models. Utilize and create larger, more diverse datasets that include a wide range of environmental conditions, object
types and scenarios to ensure that models are robust and generalize well to real-world situations. Conduct extensive
trials in various real-world environments, such as urban, rural and industrial settings, to validate and refine the models
under different conditions. Implement multi-scale feature integration techniques to better object detection of varying
sizes and improve overall detection performance.
A. REFERENCES enhancing navigation for the visually impaired.
Multimedia Tools and Applications, 2024. 83(17):
p. 52171-52195.
1. Ramisetti, C., et al. An Ultrasonic Sensor-based 15. Rajesh, P., et al. Arduino based Smart Blind Stick
blind stick analysis with instant accident alert for for People with Vision Loss. in 2023 7th
Blind People. in 2022 International Conference on International Conference on Computing
Computer Communication and Informatics Methodologies and Communication (ICCMC).
(ICCCI). 2022. IEEE. 2023. IEEE.
2. Nunes, D., et al. Real-time Vision Based Obstacle 16. Jayachitra, J., et al. Design and Implementation of
Detection in Maritime Environments. in 2022 IEEE Smart Glove for Visually Impaired People. in 2023
International Conference on Autonomous Robot 5th International Conference on Smart Systems and
Systems and Competitions (ICARSC). 2022. IEEE. Inventive Technology (ICSSIT). 2023. IEEE.
3. Assaf, E.H., et al., High-Precision Low-Cost 17. Sissodia, R., M.S. Rauthan, and V. Barthwal.
Gimballing Platform for Long-Range Railway Arduino based bluetooth voice-controlled robot car
Obstacle Detection. Sensors, 2022. 22(2): p. 474. and obstacle detector. in 2023 IEEE International
4. Shuai, Q. and X. Wu. Object detection system based Students' Conference on Electrical, Electronics and
on SSD algorithm. in 2020 international conference Computer Science (SCEECS). 2023. IEEE.
on culture-oriented science & technology (ICCST). 18. Chauhan, R., J. Upadhyay, and C. Bhatt. An
2020. IEEE. innovative wheelchair for quadreplegic patient
5. Kumar, A., Z.J. Zhang, and H. Lyu, Object using IoT. in 2023 International Conference on
detection in real time based on improved single shot Device Intelligence, Computing and
multi-box detector algorithm. EURASIP Journal on Communication Technologies,(DICCT). 2023.
Wireless Communications and Networking, 2020. IEEE.
2020: p. 1-18. 19. Lima, R., et al., Visually impaired people
6. He, D., et al., Urban rail transit obstacle detection positioning assistance system using artificial
based on Improved R-CNN. Measurement, 2022. intelligence. IEEE Sensors Journal, 2023. 23(7): p.
196: p. 111277. 7758-7765.
7. Guan, L., et al., A Lightweight Framework for 20. Prathibha, S., et al. Ultra-modern walking stick
Obstacle Detection in the Railway Image based on designed for the blind. in 2023 International
Fast Region Proposal and Improved YOLO-tiny Conference on Networking and Communications
Network. IEEE Transactions on Instrumentation (ICNWC). 2023. IEEE.
and Measurement, 2022. 71: p. 1-16. 21. Masud, U., et al., Smart assistive system for
8. Kamaruddin, F., et al. Smart Assistive Shoes with visually impaired people obstruction avoidance
Internet of Things Implementation for Visually through object detection and classification. IEEE
Impaired People. in Journal of Physics: Conference access, 2022. 10: p. 13428-13441.
Series. 2021. IOP Publishing. 22. Dhou, S., et al., An IoT machine learning-based
9. Wang, X., et al., Target Electromagnetic Detection mobile sensors unit for visually impaired people.
Method in Underground Environment: A Review. Sensors, 2022. 22(14): p. 5202.
IEEE Sensors Journal, 2022. 23. Jiang, P., et al., A Review of Yolo algorithm
10. Fang, R. and C. Cai. Computer vision based developments. Procedia Computer Science, 2022.
obstacle detection and target tracking for 199: p. 1066-1073.
autonomous vehicles. in MATEC Web of 24. Yu, T., et al., Intelligent detection method of
Conferences. 2021. EDP Sciences. forgings defects detection based on improved
11. Vorapatratorn, S. AI-Based Obstacle Detection and efficientnet and memetic algorithm. IEEE Access,
Navigation for the Blind Using Convolutional 2022. 10: p. 79553-79563.
Neural Network. in 2021 25th International 25. Arkin, E., et al., A survey: Object detection
Computer Science and Engineering Conference methods from CNN to transformer. Multimedia
(ICSEC). 2021. IEEE. Tools and Applications, 2023. 82(14): p. 21353-
12. BAMDAD, M., D. SCARAMUZZA, and A. 21383.
DARVISHY, SLAM for Visually Impaired 26. Reddy, B.S., et al. A Comparative Study on Object
Navigation: A Systematic Literature Review of the Detection Using Retinanet. in 2022 IEEE 2nd
Current State of Research. 2023. Mysore Sub Section International Conference
13. Jin, J., et al., Wide baseline stereovision based (MysuruCon). 2022. IEEE.
obstacle detection for unmanned surface vehicles. 27. Yang, Y. and J. Han, Real-Time object detector
Signal, Image and Video Processing, 2024. 18(5): based MobileNetV3 for UAV applications.
p. 4605-4614. Multimedia Tools and Applications, 2023. 82(12):
14. Patel, I., M. Kulkarni, and N. Mehendale, Review p. 18709-18725.
of sensor-driven assistive device technologies for
28. Duan, K., et al., CenterNet++ for object detection. International Journal of Computer Science and
IEEE Transactions on Pattern Analysis and Mobile Computing, 2022. 11(12): p. 99-107.
Machine Intelligence, 2023. 43. Bashir, R.N., et al., Internet of things (IoT) assisted
29. Liu, B., W. Zhao, and Q. Sun. Study of object soil salinity mapping at irrigation schema level.
detection based on Faster R-CNN. in 2017 Chinese Applied Water Science, 2022. 12(5): p. 105.
Automation Congress (CAC). 2017. IEEE. 44. Yaqoob, I. and I.S. Bajwa, Performance evaluation
30. Darms, M.S., et al., Obstacle detection and tracking of mobile stereonet for real time navigation in
for the urban challenge. IEEE Transactions on autonomous mobile robots. Multimedia Tools and
intelligent transportation systems, 2009. 10(3): p. Applications, 2024. 83(12): p. 35043-35072.
475-485. 45. Fatima, S., et al., Finding of Internal Control Genes
31. Li, L., et al. Foreground object detection from from Theileria-Infected Salivary Glands of
videos containing complex background. in Hyalomma Anatolicum Ticks by Using Real-Time
Proceedings of the eleventh ACM international Quantitative Pcr. Available at SSRN 4879539.
conference on Multimedia. 2003. 46. Bajwa, S., et al., Comparison of Outcome in Early
32. Leng, J., et al., Robust obstacle detection and vs Conventional Feeding after Colostomy Closure
recognition for driver assistance systems. IEEE in Paediatric Population. Journal of Health and
transactions on intelligent transportation systems, Rehabilitation Research, 2024. 4(1): p. 1411-1416.
2019. 21(4): p. 1560-1571. 47. Duba, P.K., N.P.B. Mannam, and P. Rajalakshmi,
33. Badrloo, S., et al., Image-based obstacle detection Stereo vision based object detection for
methods for the safe navigation of unmanned autonomous navigation in space environments.
vehicles: A review. Remote Sensing, 2022. 14(15): Acta Astronautica, 2024. 218: p. 326-329.
p. 3824. 48. Ouardi, M.M. and D.N. Jawawi, Object Detection
34. Sugimoto, S., et al. Obstacle detection using Algorithms for Autonomous Navigation
millimeter-wave radar and its visualization on Wheelchairs in Hospital Environment: Object
image sequence. in Proceedings of the 17th Detection Algorithms for Autonomous Navigation
International Conference on Pattern Recognition, Wheelchairs in Hospital Environment. International
2004. ICPR 2004. 2004. IEEE. Journal of Innovative Computing, 2024. 14(1): p. 1-
35. Devi, S.K., et al., Intelligent Deep Convolutional 6.
Neural Network Based Object Detection Model for 49. Cheng, B. and L. Deng, Vision detection and path
Visually Challenged People. Computer Systems planning of mobile robots for rebar binding. Journal
Science & Engineering, 2023. 46(3). of Field Robotics.
36. Cervera-Uribe, A.A. and P.E. Mendez-Monroy, 50. Kumar, D.N. and A. Akilandeswari. Novel
U19-Net: a deep learning approach for obstacle approach for object detection neural network model
detection in self-driving cars. Soft Computing, using you only look once v4 algorithm and
2022. 26(11): p. 5195-5207. compared with tensor flow SSD mobile net
37. Wang, X., et al. Cut and learn for unsupervised algorithm in terms of accuracy and latency. in AIP
object detection and instance segmentation. in Conference Proceedings. 2024. AIP Publishing.
Proceedings of the IEEE/CVF Conference on 51. Adiuku, N., et al., Improved Hybrid Model for
Computer Vision and Pattern Recognition. 2023. Obstacle Detection and Avoidance in Robot
38. Ikram, A., et al., Crop Yield Maximization Using Operating System Framework (Rapidly Exploring
an IoT‐Based Smart Decision. Journal of Sensors, Random Tree and Dynamic Windows Approach).
2022. 2022(1): p. 2022923. Sensors, 2024. 24(7): p. 2262.
39. Latif, M.S., et al., Pest prediction in rice using IoT 52. Zaman, M.I., et al., A robust deep networks based
and feed forward neural network. KSII multi-object multi-camera tracking system for city
Transactions on Internet and Information Systems scale traffic. Multimedia Tools and Applications,
(TIIS), 2022. 16(1): p. 133-152. 2024. 83(6): p. 17163-17181.
40. Ashiq, F., et al., CNN-based object recognition and
tracking system to assist visually impaired people.
IEEE Access, 2022. 10: p. 14819-14834.
41. Nuthakki, S., et al., Conversational AI and Llm's
Current And Future Impacts in Improving and
Scaling Health Services. International Journal of
Computer Engineering and Technology, 2023.
14(3): p. 149-155.
42. Nuthakki, S., et al., Role of AI Enabled Smart
Meters to Enhance Customer Satisfaction.