Top Research Papers on Object Detection Using YOLO

The findings from this study indicate that the developed system can effectively improve occupational safety and health management, however, there is a detection error factor caused by the lighting and colors.

The YOLOv3 (Version 3) model makes predictions with a single network evaluation, making this method extremely fast, running in real time with a capable GPU, and this implementation is going to use a simply webcam and Y OLO algorithm.

This paper aims at aiding the field of autonomous driving by helping detect objects with the use of deep learning algorithms by using state-of-the-art algorithm YOLO (you only look once) to detect different objects that appear on the road and classified into the category that they belong to with the help of bounding boxes.

A model to detect abnormal objects using the Yolov5 algorithm on thorax X-ray images was constructed using several methods to improve model accuracy: weighted boxes fusion, image transformation using Contrast Limited Adaptive Histogram (CLAHE), and data augmentation.

The You Only Look Once (YOLO) V3 technique is suggested to be used for object detection, which results more accuracy of object detection using YOLO algorithm.

This research aims to analyze the performance of image object detection methods using You Only Live Once (YOLO) specifically in the context of car detection and provide valuable insight into the use of the YOLO method in car object detection.

The evaluation results demonstrate that the mask detection system using the YOLO method achieves high detection rates and fast response times, and is expected to contribute to the effort of monitoring mask usage to control the spread of COVID-19.

The main idea is to detect and notify possible potholes without human intervention and using the YOLO algorithm, which uses a convolution neural network to detect objects in real time.

This research aimed to identify the logo on a document in the form of a college diploma by utilizing the You Only Look Once algorithm version 4 (YOLOv4) by implementing the Darknet Framework.

-This paper spotlight on the real time detection and recognition model called as YOLO. It uses single convolutional neural network in order to detect and recognize the objects of the images. The model is first trained on COCO dataset and car dataset of achieving a mAP of 91.28% and 70% respectively. YOLO takes 57 FPS to processes the image to detect the objects in Image. Since YOLO takes whole detection pipeline in a single unified network and it helps to increase and optimize the real time object detection with variety of objects.

Future research needs to focus on developing adaptive YOLO architectures capable of handling diverse spectral inputs that do not require extensive architectural modifications, and innovating fusion research with other sensor types beyond RGB and LWIR.

The combination of two YOLOs and ANN resulted in more successful object detection than relying solely on one YOLO, even in cases where only one YOLO struggled to detect an object or made incorrect detections.

Concept Neural Network (CNN) is employed in YOLO (You Only Look Once) algorithm to detect Objects in real-time to achieve the requirements of latest object detection of the advanced world.

It can be concluded that the goodie bag detection model has been successfully created and it can be said that YOLO v5 can detect goodie bags very well.

This work highlights some of the future research directions for YOLO variants, such as improving their robustness to different environmental conditions like motion blur, lighting condition and integrating them with other computer vision tasks like image segmentation, image classification and object tracking.

Natural disasters are events that cannot be predicted both by location and time of occurrence. Natural disasters cause property losses and can even take lives. The handling of rapid evacuation must be done by the SAR team to help victims of natural disasters to reduce the amount of loss. But in reality, there are many obstacles in the evacuation process. Starting from facing difficult terrain to necessary equipment limitation. In this research, a system designed to detect victims of natural disasters uses image processing where the picture is carried out using a drone that aims to help find vi...

This paper presents a novel approach to underwater object detection using the YOLOV5 algorithm, which leverages the capabilities to efficiently and accurately identify and localize objects in underwater imagery.

The study critically assesses and summarizes the information on YOLO's end-to-end learning approach, including data acquisition, processing, network modification, integration, and deployment, and discussed task-specific YOLO algorithm modification and integration to meet the agricultural object or environment-specific challenges.

YOLO Nano is introduced, a highly compact deep convolutional neural network for the task of object detection that combines principled network design prototyping, based on design principles from the YOLO family of single-shot object detection network architectures, with machine-driven design exploration to create a compact network with highly customized module-level macroarchitectures and microarchitecture designs tailored for thetask of embedded object detection.

The proposed Novel You Only Look Once algorithm seems to be significantly better than the tiny-yolo algorithm at detecting, recognizing, and tracking with more accuracy.

The developed algorithms implement the YOLO model for counting vehicles in two-way directions in an efficient way and the accuracy of the automated counting was evaluated compared to the manual counts, and was found to be about 90 percent.

The preliminary YOLO test results showed an increase in the accuracy of ship detection at Cyprus’s Coast and can be applied in the field of ship Detection.

Compared to state-of-the-art detection systems, YOLO makes more localization errors but is less likely to predict false positives on background, and outperforms other detection methods, including DPM and R-CNN, when generalizing from natural images to other domains like artwork.

A system to develop a third eye-hand glove object detection for visually challenged people with the You Only Look Once (YOLO)v4-tiny algorithm that detects indoor objects with the overall F1 score, precision, recall, and accuracy is 83.00%.

The evolutionary deep intelligence framework is leveraged to evolve the YOLOv2 network architecture and produce an optimized architecture that has 2.8X fewer parameters with just a ~2% IOU drop, and a motion-adaptive inference method is introduced into the proposed Fast Y OLO framework to reduce the frequency of deep inference with O-YOLO v2 based on temporal motion characteristics.

Aerial images have distinct characteristics, such as varying target scales, complex backgrounds, severe occlusion, small targets, and dense distribution. As a result, object detection in aerial images faces challenges like difficulty in extracting small target information and poor integration of spatial and semantic data. Moreover, existing object detection algorithms have a large number of parameters, posing a challenge for deployment on drones with limited hardware resources. We propose an efficient small-object YOLO detection model (ESOD-YOLO) based on YOLOv8n for Unmanned Aerial Vehicle (U...

This research focuses on "You Only Look Once" (YOLO) as a type of Convolutional Neural Network as a method to detect objects in images and video using Yolo3-tiny.

This work proposes a visual-based detector to reduce the risk of blind spots in causing road injuries to motorcyclists in Malaysia and shows that YOLO detector is the most superior detector since it has the best average precision out of all detectors.

This study proposes a system that employs the you only look once (YOLO) algorithm to detect various objects in the proximity of a robot and enhances human–robot interaction capabilities via data transmission.

The authors address two main challenges of a computer vision system for unmanned aerial vehicles in utility asset inspection: the scarcity of data and the detection of six power line components.

This research aims at detecting objects for indoor environment such as offices or rooms in different conditions of lighting by using YOLOv3 and generating a voice message for each detected object by using YOLOv3.

A novel MLaaS-based system that partitions 360° frames into distortion-free 2D regions with dynamic region of interest prediction and seamlessly combining all the 2D regions into a unified frame, proving its effectiveness in 360° video object detection tasks.

A thorough review of the application of YOLO in smoke and fire detection over the previous three years using data sets, methodology, strategies and evaluating performance is given.

The first findings of vehicle route accounting reveal that the technique is highly promising, with good outcomes in a number of scenarios, but much more study is needed to make these systems resistant against occlusions and other unforeseen events.

The body detection system of natural disaster victims is designed using image processing where the shooting of victims was carried out using drones aiming to help find victims in a difficult or prone location when reached directly by humans.

Analysis of the ability of a machine learning framework named “You Only Look Once,” to perform object localization task in a “Heuristic once learning” context showed that YOLO had difficulties to generalize simple abstractions of the characters, pointing to the necessity of new approaches to solve such challenges.

This research was undertaken by applying the You Only Look Once (yolo) algorithm with several test scenarios to see the performance generated by the system, because yolo is one of the fastest and most accurate methods for object detection and even exceeds 2 times the capabilities of other algorithms.

A new dataset whose images were taken by a web camera from a jewellery store and data augmentation procedure is introduced and it comprises three classes, namely, ring, earrings, and pendant, which are focused on small object detection especially jewellery.

A novel three- dimensional object detection method based on two-dimensional object detection, which only takes a set of RGB images as input and can effectively realize the three- dimension object detection without depth images.

The new loss function is introduced in the YOLO v5 which improves the performance of object detection and the CIoU-YOLO v5 method attained the Average Precision (AP) of 44.7% on MS-COCO 2017 dataset and 53.7% AP on KITTI dataset which is superior than Dual-path Lightweight Module (DLM).

This article employed the YOLO family models and trained them using the transfer learning approach on custom datasets, and found that these cutting- edge performers outperformed the competition on the Custom Dataset of Open Image Datasets V6 for various classes.

The algorithm presented in this work gains the performance speedup used in previous motion based neural network inference papers and also performs a novel look back operation that is less cumbersome than other competing motion interpolation methods.

Experimental results demonstrate that using auto-generated labels for object detection does not lead to degradation in performance and the efficacy of auto-labeling technology in contributing to efficiency and performance enhancement in the field of object detection, presenting practical applicability.

It was proved that the novel you only look once algorithm has a greater accuracy than deep neural networks.

The correct ?r?blem ?

YOLO, a way to deal with item recognition, is demonstrated and Convolutional Neural Network or CNN, is a method which has had the capacity to effectively take care of the picture acknowledgment issue productively.

This paper presents a survey of various detections based on YOLO which aims to improve the accuracy of existing system and presents various modifications done on basic Y OLO method and shows their analysis.

The YOLO algorithm model detects and recognizes the objects, and the improved model will examine the entire image and split the image into regions and maps the confidence probability using a neural network on the image.

The growing need for automated object detection and recognition in visual data has become crucial in various domains, including surveillance, autonomous vehicles, and image analysis. Traditional methods often struggle with speed and accuracy, making real-time applications challenging. The main objective of this experiment is to make the computers identify and detect objects in the visual data. The algorithm is used to detect the object with classification and localization based on visual inputs. This project uses YOLO(You Only Look Once) algorithm to detect the objects. YOLO v8 is known for it...

Investigation of fire and smoke objects in the form of design, implementation and testing resulted in the YOLOv4 framework successfully producing high confidence of up to 97% in the second test, known that the image datasets used for training data greatly affect object detection and affect the confidence value.