IDOD-YOLOV7: Image-Dehazing YOLOV7 for Object Detection in Low-Light Foggy Traffic Environments

Abstract

Convolutional neural network (CNN)-based autonomous driving object detection algorithms have excellent detection results on conventional datasets, but the detector performance can be severely degraded in low-light foggy weather environments. Existing methods have difficulty in achieving a balance between low-light image enhancement and object detection. To alleviate this problem, this paper proposes a foggy traffic environment object detection framework, IDOD-YOLOV7. This network is based on joint optimal learning of image defogging module IDOD (AOD + SAIP) and YOLOV7 detection modules. Specifically, for low-light foggy images, we propose to improve the image quality by joint optimization of image defogging (AOD) and image enhancement (SAIP), where the parameters of the SAIP module are predicted by a miniature CNN network and the AOD module performs image defogging by optimizing the atmospheric scattering model. The experimental results show that the IDOD module not only improves the image defogging quality for low-light fog images but also achieves better results in objective evaluation indexes such as PSNR and SSIM. The IDOD and YOLOV7 learn jointly in an end-to-end manner so that object detection can be performed while image enhancement is executed in a weakly supervised manner. Finally, a low-light fogged traffic image dataset (FTOD) was built by physical fogging in order to solve the domain transfer problem. The training of IDOD-YOLOV7 network by a real dataset (FTOD) improves the robustness of the model. We performed various experiments to visually and quantitatively compare our method with several state-of-the-art methods to demonstrate its superiority over the others. The IDOD-YOLOV7 algorithm not only suppresses the artifacts of low-light fog images and improves the visual effect of images but also improves the perception of autonomous driving in low-light foggy environments.

DCP-YOLOv7: dark channel prior image-dehazing YOLOv7 for vehicle detection in foggy scene

“…McCartney introduced an atmospheric scattering model [5] to elucidate the formation process of fog images as shown in Figure 2. In accordance with this model, fog is formed due to the absorption and scattering of natural light caused by the high density of water vapor and minute suspended particles in the atmosphere [13] . As the particles disperse the light, the transmission between the object and the sensor results in a decrease in light intensity, and an extra layer of scattered atmospheric light is added.…”

A military reconnaissance network for small‐scale open‐scene camouflaged people detection

“…Following the current metric for object detection (Lin et al, 2022), we primarily evaluate the detection mean Average Precision (mAP), frames per second (FPS). In this paper, we compare SM-CODN with other state-of-the-art object detection methods, including SSD (Liu et al, 2016), DETR (Carion et al, 2020), YOLOv7 (Wang, Bochkovskiy, et al, 2022), PPYOLOE-M (Shangliang et al, 2022), AdaMixer (Gao et al, 2022), MCS-YOLO (Shu-Jun et al, 2023), IDOD-YOLOV7 (Qiu et al, 2023), RF-Next (Gao et al, 2023). Our model is based on the advanced YOLOv7 which has a good trade-off between accuracy and inference speed.…”

Enhancing Object Recognition in Hazy Scenes with Boundary-Constrained Dehazing and YOLOv7 Architecture

“…The YOLO (You Only Look Once) architecture is a popular approach for object detection, which uses a single convolutional neural network (CNN) to predict bounding boxes and class probabilities. Several versions of YOLO have been proposed, including YOLOv2, YOLOv3, and YOLOv4 [11][14] [17]. In this paper, YOLOv7 architecture is proposed.The COCO (Common Objects in Context) dataset is a commonly used dataset for object detection, which contains over 330,000 images and more than 2.5 million object instances [18].…”

“…However, they all aim to improve object detection accuracy in hazy scenes by improving visibility and restoring image quality through image dehazing. The results in the table suggest that boundary-constrained dehazing combined with Faster R-CNN [19] and YOLOv3 [14] architecture are effective approaches for improving object detection accuracy in hazy scenes, as they achieve high mAP scores and relatively high SSIM and PSNR values [11].…”

“…However, the performance of these algorithms in hazy scenes can be signi cantly degraded [10]. The YOLOv7 architecture uses a convolutional neural network (CNN) for object detection, which can be represented mathematically as [11]:…”