Vehicle Target Detection in Rainy and Foggy Scenes Based on Generative Adversarial Networks and Dynamic Fuzzy Compensation Techniques

With the rapid development in the field of artificial intelligence and the advancement of deep learning theory, vehicle target detection technology has been widely used in the field of urban intelligent transportation and automatic driving, assisting vehicles to achieve safe driving in complex driving environments and improving traffic safety. This paper proposes a dynamic fuzzy image processing method based on Wiener filter and generative adversarial network, and constructs a UNIT-based de-fogging and de-raining algorithm, which can be generalized to clarify the targets obtained in rainy and foggy scenes. Then design the local perception enhancement vehicle detection model assisted by image rain removal to realize the accurate detection of vehicle targets in rainy and foggy scenes. By applying the method of this paper on the synthetic dataset Rain Vehicle Color-24, the results demonstrate that the mAP values of this paper’s method are 3.73%, 2.23% and 1.19% higher than those of Da-Faster, SA-Da-Faster and SMNN-MSFF respectively, which are able to improve the vehicle color recognition task in rainy and foggy scenes with good Accuracy. Therefore, the method in this paper can reduce the domain differences of the model in the target domain and improve the localization accuracy.