No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Performance Evaluation of YOLOv5 in Adverse Weather Conditions
... The performance evaluation was performed on the DAWN dataset using only the YOLO-V5n model. They only focused on domain-shift analysis and the effect of transfer learning without providing vehicle detection results and analysis [27]. Another study focused on performance evaluation on Roboflow datasets using only the YOLO-V5s model. ...
Vision-based vehicle detection in adverse weather conditions such as fog, haze, and mist is a challenging research area in the fields of autonomous vehicles, collision avoidance, and Internet of Things (IoT)-enabled edge/fog computing traffic surveillance and monitoring systems. Efficient and cost-effective vehicle detection at high accuracy and speed in foggy weather is essential to avoiding road traffic collisions in real-time. To evaluate vision-based vehicle detection performance in foggy weather conditions, state-of-the-art Vehicle Detection in Adverse Weather Nature (DAWN) and Foggy Driving (FD) datasets are self-annotated using the YOLO LABEL tool and customized to four vehicle detection classes: cars, buses, motorcycles, and trucks. The state-of-the-art single-stage deep learning algorithms YOLO-V5, and YOLO-V8 are considered for the task of vehicle detection. Furthermore, YOLO-V5s is enhanced by introducing attention modules Convolutional Block Attention Module (CBAM), Normalized-based Attention Module (NAM), and Simple Attention Module (SimAM) after the SPPF module as well as YOLO-V5l with BiFPN. Their vehicle detection accuracy parameters and running speed is validated on cloud (Google Colab) and edge (local) systems. The mAP50 score of YOLO-V5n is 72.60%, YOLO-V5s is 75.20%, YOLO-V5m is 73.40%, and YOLO-V5l is 77.30%; and YOLO-V8n is 60.20%, YOLO-V8s is 73.50%, YOLO-V8m is 73.80%, and YOLO-V8l is 72.60% on DAWN dataset. The mAP50 score of YOLO-V5n is 43.90%, YOLO-V5s is 40.10%, YOLO-V5m is 49.70%, and YOLO-V5l is 57.30%; and YOLO-V8n is 41.60%, YOLO-V8s is 46.90%, YOLO-V8m is 42.90%, and YOLO-V8l is 44.80% on FD dataset. The vehicle detection speed of YOLO-V5n is 59 Frame Per Seconds (FPS), YOLO-V5s is 47 FPS, YOLO-V5m is 38 FPS, and YOLO-V5l is 30 FPS; and YOLO-V8n is 185 FPS, YOLO-V8s is 109 FPS, YOLO-V8m is 72 FPS, and YOLO-V8l is 63 FPS on DAWN dataset. The vehicle detection speed of YOLO-V5n is 26 FPS, YOLO-V5s is 24 FPS, YOLO-V5m is 22 FPS, and YOLO-V5l is 17 FPS; and YOLO-V8n is 313 FPS, YOLO-V8s is 182 FPS, YOLO-V8m is 99 FPS, and YOLO-V8l is 60 FPS on FD dataset. YOLO-V5s, YOLO-V5s variants and YOLO-V5l_BiFPN, and YOLO-V8 algorithms are efficient and cost-effective solution for real-time vision-based vehicle detection in foggy weather.
... Resilience and robustness despite unfavorable weather conditions surface as critical challenges in ITS. The inconsistently changing weather behavior presents a substantial obstacle to the reliable implementation of ITS operations, impacting sensor data quality, communication channels, and the system's operational efficiency [262]. Downpours, snow, dense fog, and other adverse weather conditions can substantially affect the performance of DL models and traditional ML algorithms deployed in transportation systems [263]. ...
... Resilience and robustness despite unfavorable weather conditions surface as critical challenges in ITS. The inconsistently changing weather behavior presents a substantial obstacle to the reliable implementation of ITS operations, impacting sensor data quality, communication channels, and the system's operational efficiency [262]. Downpours, snow, dense fog, and other adverse weather conditions can substantially affect the performance of DL models and traditional ML algorithms deployed in transportation systems [263]. ...
Intelligent Transportation Systems (ITS) operate within a highly intricate and dynamic environment characterized by complex spatial and temporal dynamics at various scales, further compounded by fluctuating conditions influenced by external factors such as social events, holidays, and weather conditions. Navigating the intricacies of modeling the intricate interaction among these elements, creating universal representations, and employing them to address particular transportation issues presents a significant endeavor. Yet, these intricacies comprise just one facet of the multifaceted trials confronting contemporary ITS. This paper offers an all-encompassing survey that explores Deep learning (DL) utilisation in ITS, primarily focusing on the methodologies practitioners adopt in addressing these multifaceted challenges. The emphasis lies on the architectural and problem-specific factors that guide the formulation of innovative solutions. In addition to shedding light on the state-of-the-art DL algorithms, we also explore potential applications of DL in ITS, including traffic flow prediction, vehicle detection and classification, road condition monitoring, traffic sign recognition, and autonomous vehicles. Besides, we identify several future challenges and research directions that can push the boundaries of ITS, including the critical aspects of explainability, transfer learning, hybrid models, privacy and security, and ultra-reliable low-latency communication. Our aim for this survey is to bridge the gap between the burgeoning DL and transportation communities. By doing so, we aim to facilitate a deeper comprehension of the challenges and possibilities within this field. We hope that this effort will inspire further exploration of fresh perspectives and issues, which, in turn, will play a pivotal role in shaping the future of Intelligent Transportation Systems.
Unsupervised domain adaptation (UDA) is a transfer learning technique utilized in deep learning. UDA aims to reduce the distribution gap between labeled source and unlabeled target domains by adapting a model through fine-tuning. Typically, UDA approaches assume the same categories in both domains. The effectiveness of transfer learning depends on the degree of similarity between the domains, which determines an efficient fine-tuning strategy. Furthermore, domain-specific tasks generally perform well when the feature distributions of the domains are similar. However, utilizing a trained source model directly in the target domain may not generalize effectively due to domain shift. Domain shift can be caused by intra-class variations, camera sensor variations, background variations, and geographical changes. To address these issues, we design an efficient unsupervised domain adaptation network for image classification and object detection that can learn transferable feature representations and reduce the domain shift problem in a unified network. We propose the guided transfer learning approach to select the layers for fine-tuning the model, which enhances feature transferability and utilizes the JS-Divergence to minimize the domain discrepancy between the domains. We evaluate our proposed approaches using multiple benchmark datasets. Our domain adaptive image classification approach achieves 93.2% accuracy on the Office-31 dataset and 75.3% accuracy on the Office-Home dataset. In addition, our domain adaptive object detection approach achieves 51.1% mAP on the Foggy Cityscapes dataset and 72.7% mAP on the Indian Vehicle dataset. We conduct extensive experiments and ablation studies to demonstrate the effectiveness and efficiency of our work. Experimental results also show that our work significantly outperforms the existing methods.
Drones/unmanned aerial vehicles (UAVs) have recently grown in popularity due to their inexpensive cost and widespread commercial use. The increased use of drones raises the possibility that they may be employed in illicit activities such as drug smuggling and terrorism. Thus, drone monitoring and automated detection are critical for protecting restricted areas or special zones from illicit drone operations. One of the most challenging difficulties in drone detection in surveillance videos is the apparent likeness of drones against varied backdrops. This paper introduces an automated image-based drone-detection system that uses an enhanced deep-learning-based object-detection algorithm known as you only look once (YOLOv5) to defend restricted territories or special zones from unauthorized drone incursions. The transfer learning to pretrain the model is employed for improving performance due to an insufficient number of samples in our dataset. Furthermore, the model can recognize the detected object in the images and mark the object’s bounding box by joining the results across the region. The experiments show outstanding results for the loss value, drone location detection, precision and recall.
Recent progress in deep learning has led to accurate and efficient generic object detection networks. Training of highly reliable models depends on large datasets with highly textured and rich images. However, in real-world scenarios, the performance of the generic object detection system decreases when (i) occlusions hide the objects, (ii) objects are present in low-light images, or (iii) they are merged with background information. In this paper, we refer to all these situations as challenging environments. With the recent rapid development in generic object detection algorithms, notable progress has been observed in the field of deep learning-based object detection in challenging environments. However, there is no consolidated reference to cover the state of the art in this domain. To the best of our knowledge, this paper presents the first comprehensive overview, covering recent approaches that have tackled the problem of object detection in challenging environments. Furthermore, we present a quantitative and qualitative performance analysis of these approaches and discuss the currently available challenging datasets. Moreover, this paper investigates the performance of current state-of-the-art generic object detection algorithms by benchmarking results on the three well-known challenging datasets. Finally, we highlight several current shortcomings and outline future directions.
With the rapid development of deep learning techniques, deep convolutional neural networks (DCNNs) have become more important for object detection. Compared with traditional handcrafted feature-based methods, the deep learning-based object detection methods can learn both low-level and high-level image features. The image features learned through deep learning techniques are more representative than the handcrafted features. Therefore, this review paper focuses on the object detection algorithms based on deep convolutional neural networks, while the traditional object detection algorithms will be simply introduced as well. Through the review and analysis of deep learning-based object detection techniques in recent years, this work includes the following parts: backbone networks, loss functions and training strategies, classical object detection architectures, complex problems, datasets and evaluation metrics, applications and future development directions. We hope this review paper will be helpful for researchers in the field of object detection.
Recent work in domain adaptation bridges different domains by adversarially learning a domain-invariant representation that cannot be distinguished by a domain discriminator. Existing methods of adversarial domain adaptation mainly align the global images across the source and target domains. However, it is obvious that not all regions of an image are transferable, while forcefully aligning the untransferable regions may lead to negative transfer. Furthermore, some of the images are significantly dissimilar across domains, resulting in weak image-level transferability. To this end, we present Transferable Attention for Domain Adaptation (TADA), focusing our adaptation model on transferable regions or images. We implement two types of complementary transferable attention: transferable local attention generated by multiple region-level domain discriminators to highlight transferable regions, and transferable global attention generated by single image-level domain discriminator to highlight transferable images. Extensive experiments validate that our proposed models exceed state of the art results on standard domain adaptation datasets.
Object detection typically assumes that training and test data are drawn from an identical distribution, which, however, does not always hold in practice. Such a distribution mismatch will lead to a significant performance drop. In this work, we aim to improve the cross-domain robustness of object detection. We tackle the domain shift on two levels: 1) the image-level shift, such as image style, illumination, etc, and 2) the instance-level shift, such as object appearance, size, etc. We build our approach based on the recent state-of-the-art Faster R-CNN model, and design two domain adaptation components, on image level and instance level, to reduce the domain discrepancy. The two domain adaptation components are based on H-divergence theory, and are implemented by learning a domain classifier in adversarial training manner. The domain classifiers on different levels are further reinforced with a consistency regularization to learn a domain-invariant region proposal network (RPN) in the Faster R-CNN model. We evaluate our newly proposed approach using multiple datasets including Cityscapes, KITTI, SIM10K, etc. The results demonstrate the effectiveness of our proposed approach for robust object detection in various domain shift scenarios.
Machine learning and data mining techniques have been used in numerous real-world applications. An assumption of traditional machine learning methodologies is the training data and testing data are taken from the same domain, such that the input feature space and data distribution characteristics are the same. However, in some real-world machine learning scenarios, this assumption does not hold. There are cases where training data is expensive or difficult to collect. Therefore, there is a need to create high-performance learners trained with more easily obtained data from different domains. This methodology is referred to as transfer learning. This survey paper formally defines transfer learning, presents information on current solutions, and reviews applications applied to transfer learning. Lastly, there is information listed on software downloads for various transfer learning solutions and a discussion of possible future research work. The transfer learning solutions surveyed are independent of data size and can be applied to big data environments.
Transportation or transport sector is a legal source to take or carry things from one place to another. With the passage of time, transportation faces many issues like high accidents rate, traffic congestion, traffic & carbon emissions air pollution, etc. In some cases, transportation sector faced alleviating the brutality of crash related injuries in accident. Due to such complexity, researchers integrate virtual technologies with transportation which known as Intelligent Transport System. The idea of virtual technologies integration is a novel in transportation field and it plays a vital part to overcome the issues in global world. This paper tackles the great variety of Intelligent Transport System applications, technologies and its different areas. The objective of this literature review is to integrate and synthesize some areas and applications, technologies discuss with all prospects. Furthermore, this research focuses on a wide field named Intelligent Transport Systems, discussed its wide applications, used technologies and its usage in different areas respectively.
Adverse weather conditions such as haze and rain corrupt the quality of captured images, which cause detection networks trained on clean images to perform poorly on these corrupted images. To address this issue, we propose an unsupervised prior-based domain adversarial object detection framework for adapting the detectors to hazy and rainy conditions. In particular, we use weather-specific prior knowledge obtained using the principles of image formation to define a novel prior-adversarial loss. The prior-adversarial loss, which we use to supervise the adaptation process, aims to reduce the weather-specific information in the features, thereby mitigating the effects of weather on the detection performance. Additionally, we introduce a set of residual feature recovery blocks in the object detection pipeline to de-distort the feature space, resulting in further improvements. Evaluations performed on various datasets (Foggy-Cityscapes, Rainy-Cityscapes, RTTS and UFDD) for rainy and hazy conditions demonstrates the effectiveness of the proposed approach.
Due to object detection's close relationship with video analysis and image understanding, it has attracted much research attention in recent years. Traditional object detection methods are built on handcrafted features and shallow trainable architectures. Their performance easily stagnates by constructing complex ensembles that combine multiple low-level image features with high-level context from object detectors and scene classifiers. With the rapid development in deep learning, more powerful tools, which are able to learn semantic, high-level, deeper features, are introduced to address the problems existing in traditional architectures. These models behave differently in network architecture, training strategy, and optimization function. In this paper, we provide a review of deep learning-based object detection frameworks. Our review begins with a brief introduction on the history of deep learning and its representative tool, namely, the convolutional neural network. Then, we focus on typical generic object detection architectures along with some modifications and useful tricks to improve detection performance further. As distinct specific detection tasks exhibit different characteristics, we also briefly survey several specific tasks, including salient object detection, face detection, and pedestrian detection. Experimental analyses are also provided to compare various methods and draw some meaningful conclusions. Finally, several promising directions and tasks are provided to serve as guidelines for future work in both object detection and relevant neural network-based learning systems.
We present a new dataset with the goal of advancing the state-of-the-art in
object recognition by placing the question of object recognition in the context
of the broader question of scene understanding. This is achieved by gathering
images of complex everyday scenes containing common objects in their natural
context. Objects are labeled using per-instance segmentations to aid in
understanding an object's precise 2D location. Our dataset contains photos of
91 objects types that would be easily recognizable by a 4 year old along with
per-instance segmentation masks. With a total of 2.5 million labeled instances
in 328k images, the creation of our dataset drew upon extensive crowd worker
involvement via novel user interfaces for category detection, instance spotting
and instance segmentation. We present a detailed statistical analysis of the
dataset in comparison to PASCAL, ImageNet, and SUN. Finally, we provide
baseline performance analysis for bounding box and segmentation detection
results using a Deformable Parts Model.
Microsoft Coco: Common Objects in Context
- lin
DAWN: Vehicle Detection in Adverse Weather Nature Dataset
- M A Kenk
- M Hassaballah
YOLOv5 by Ultralytics (Version 7.0) [Computer software]
- G Jocher
Detection of unauthorized unmanned aerial vehicles using Yolov5 and transfer learning
- N Al-Qubaydhi
- A Alenezi
- T Alenezi
- A Senyor
- N Alanezi
- B Alotaibi
- M Alotaibi
- A Razaque
- A A Abdelhamid
- A Alotaibi
In Rain or Shine: Understanding and Overcoming Dataset Bias for Improving Robustness Against Weather Corruptions for Autonomous Vehicles
- A Marathe
- R Walambe
- K Kotecha
Microsoft Coco: Common Objects in Context
- T.-Y Lin
- M Maire
- S Belongie
- J Hays
- L Bourdev
- R Girshick