Search

An autonomous vehicle must accurately detect its surrounding environment to operate reliably. Adverse weather conditions (ADWC) are snow, rain, sand, and haze, badly affect the quality of vehicle detection (VD) in an autonomous environment. Most existing techniques focused on VD under various weather effects such as signal control, travel pattern, traffic volume variations and collision risk. Only a limited number of works of literature were focused on VD under ADWC at different automation scales. In this paper, a novel deep learning (DL) model, Optimized You Look Only Once Version 5 (OYOLOV5), is proposed for autonomous VD (AVD) in ADWC. The proposed model consists of three phases: data collection, data preprocessing, feature extraction and classification. Initially, the data is collected from the DAWN and COCO dataset to perform VD, which is openly available. The augmentation of the data is carried out on the collected input data by including hue, saturation, blur, brightness, and noise, which helps to get a clear view of vehicles. After data augmentation, feature extraction and classification of the preprocessed images are done using the OYOLOV5 framework, which uses Resnet-50 as the backbone network and Feature Pyramid Network (FPN) for detecting the vehicles at multi-scales. Experiments are conducted, and the outcomes demonstrated the proposed OYOLOV5 model achieves better performance with the state-of-art methods in terms of precision (PRC), recall (RC), f-measure (FMS), accuracy (ACU), average IoU (AI), processing speed (PS), and training time (TTI). Also, the system attains good mean average precision (mAP) than the conventional methods. [ABSTRACT FROM AUTHOR]

Automated object detection has received the most attention over the years. Use cases ranging from autonomous driving applications to military surveillance systems, require robust detection of objects in different illumination conditions. State-of-the-art object detectors tend to fare well in object detection during daytime conditions. However, their performance is severely hampered in night light conditions due to poor illumination. To address this challenge, the manuscript proposes an improved YOLOv5-based object detection framework for effective detection in unevenly illuminated nighttime conditions. Firstly, the preprocessing strategies involve using the Zero- DCE++ approach to enhance lowlight images. It is followed by optimizing the existing YOLOv5 architecture by integrating the Convolutional Block AttentionModule (CBAM) in the backbone network to boost model learning capability and Depthwise Convolutional module (DWConv) in the neck network for efficient compression of network parameters. The Night Object Detection (NOD) and Exclusively Dark (ExDARK) dataset has been used for this work. The proposed framework detects classes like humans, bicycles, and cars. Experiments demonstrate that the proposed architecture achieved a higher Mean Average Precision (mAP) along with a reduction in model size and total parameters, respectively. The proposedmodel is lighter by 11.24% in terms ofmodel size and 12.38% in terms of parameters when compared to baseline YOLOv5. [ABSTRACT FROM AUTHOR]