Comparison of YOLO v3, Faster R-CNN, and SSD for Real-Time Pill Identification

Tan, Lu; Huangfu, Tianran; Wu, Liyao; Chen, Wenying

doi:10.21203/rs.3.rs-668895/v1

Cited by 18 publications

(5 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many studies with segmentation have improved object detection performance, but the accuracy still stays around 80% [5][6][7][8][9][10]. The accuracy of most segmentation algorithms is higher than Yolo algorithms', but the efficiency is much worse [11][12][13].…”

Section: Related Workmentioning

confidence: 99%

Hierarchical Structure for Semantic Segmentation of Roadway Images with Limited Visibility Using Deep Artificial Neural Networks

An,

Mohan,

Leppo

et al. 2023

Preprint

View full text Add to dashboard Cite

The field of autonomous driving leaves minimal margins for error. Ensuring that self-driving vehicles possess the ability to accurately perceive their surroundings, even amidst conditions of limited visibility, is of utmost importance. We propose a novel approach to enhance the precision of object detection on the road during limited visibility driving or roadway conditions. The initial step involves the classification of the driving condition of an input image, and then, the corresponding semantic segmentation model will process the image to distinguish objects. Our dataset consists of roadway images depicting 20 distinct objects amidst adverse limited visibility conditions. The experimental results validate our approach, with the proposed method displaying quality accuracy levels for training, validation, and testing data. Our classification model achieved 100% accuracy. Particularly, the proposed methods achieved final mean IoU scores of 57.3%, 32.0%, 49.4%, and 47.8%, respectively, for FOG, NIGHT, RAIN, and SNOW conditions when using the U-NET model for segmentation. These mean IoU results are better than the traditional nonhierarchical training methods, which utilize the same U-NET structure.

show abstract

Section: Related Workmentioning

confidence: 99%

Hierarchical Structure for Semantic Segmentation of Roadway Images with Limited Visibility Using Deep Artificial Neural Networks

An,

Mohan,

Leppo

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…While most algorithms repurpose classifiers by taking images at multiple scales and locations and applying the algorithm to perform detection [13], YOLO applies a neural network that dissects an image into different parts and can predict bounding box regions based on predicted probabilities [14]. YOLO detects objects using a single inference which makes it faster than its peers, SSD (Single Shot Detector) and Faster R-CNN (Region-based Convolutional Neural Network) [15].…”

Section: Literature Reviewmentioning

confidence: 99%

Performance Analysis of Deep Learning YOLO models for South Asian Regional Vehicle Recognition

Rafi¹,

Chakma²,

Mahmud³

et al. 2022

IJACSA

View full text Add to dashboard Cite

For years, humans have pondered the possibility of combining human and machine intelligence. The purpose of this research is to recognize vehicles from media and while there are multiple models associated with this, models that can detect vehicles commonly used in developing countries like Bangladesh, India, etc. are scarce. Our focus was to assimilate the largest dataset of vehicles exclusive to South Asia in addition to the more common universal vehicles and apply it to track and recognize these vehicles, even in motion. To develop this, we increased the class variations and quantity of the data and used multiple variations of the YOLOv5 model. We trained different versions of the model with our dataset to properly measure the degree of accuracy between the models in detecting the more unique vehicles. If vehicle detection and tracking are adopted and implemented in live traffic camera feeds, the information can be used to create smart traffic systems that can regulate congestion and routing by identifying and separating fast and slow-moving vehicles on the road. The comparison between the three different YOLOv5 models led to an analysis that indicates that the large variant of the YOLOv5 architecture outperforms the rest.

show abstract

“…Moreover, this architecture is more powerful for identifying even small objects from images [15]. YOLO v3 architecture also has advantages in detection speed while maintaining a specific Mean Average Precision (mAP), is relatively easy to modify, and has a faster computation time [16].…”

Section: Introductionmentioning

confidence: 99%

Development of Automatic Counting System for Palm Oil Tree Based on Remote Sensing Imagery

Muna¹,

Nugroho²,

Syarovy³

et al. 2022

Advances in Biological Sciences Research

View full text Add to dashboard Cite

Data on the number of palm oil tree plantations on cultivated land is essential in a company's cultivation activities. Limitations of collecting data number of palm oil trees using the terrestrial method are the effectiveness of times, in terms of costs, and coverage area. Utilization of remote sensing with aerial imagery and deep learning method could present the results more efficiently. This research aims to detect and calculate the number of palm oil trees using the You Only Look Once (YOLO) version 3 architecture object detection model based on remote sensing imagery. The aerial image is collected using the Unmanned Aerial Vehicle (UAV) to train and validation the model. The detection results by the model are stored as a shapefile for further processing using the Quantum Geographic Information System (Q-GIS) to determine the number and display the detection results of palm oil trees. The total number of objects detected as trees through the model is 559 palm oil trees. The actual number of palm oil trees recorded was 590 palm oil trees. Based on the Mean Average Percentage Error (MAPE) value obtained, which is 0.057627, it shows that the model built is good and can be used to estimate the number of palm oil trees. In the future, evaluation and optimization of the model can be carried out by adjusting the number of iterations and increasing the amount of training data.

show abstract

Comparison of YOLO v3, Faster R-CNN, and SSD for Real-Time Pill Identification

Cited by 18 publications

References 23 publications

Hierarchical Structure for Semantic Segmentation of Roadway Images with Limited Visibility Using Deep Artificial Neural Networks

Hierarchical Structure for Semantic Segmentation of Roadway Images with Limited Visibility Using Deep Artificial Neural Networks

Performance Analysis of Deep Learning YOLO models for South Asian Regional Vehicle Recognition

Development of Automatic Counting System for Palm Oil Tree Based on Remote Sensing Imagery

Contact Info

Product

Resources

About