A Novel Algorithm for Detecting Pedestrians on Rainy Image

Abstract

Pedestrian detection is widely used in cooperative vehicle infrastructure systems. Traditional pedestrian detection methods perform sufficiently well under sunny scenarios and obtain trustworthy traffic data. However, the detection drastically decreases under rainy scenarios. This study proposes a pedestrian detection algorithm with a de-raining module that improves detection accuracy under various rainy scenarios. Specifically, this algorithm determines the density information of rain and effectively removes rain streaks through the de-raining module. Then the algorithm detects pedestrians as a pair of keypoints through the pedestrian detection module to solve the problem of occlusion. Furthermore, a new pedestrian dataset containing rain density labels is established and used to train the algorithm. For the scenarios of light, medium, and heavy rain, extensive experiments on synthetic datasets demonstrate that the proposed algorithm increases AP (average precision) of pedestrian detection by 21.1%, 48.1%, and 60.9%. Moreover, the proposed algorithm performs well on real datasets and achieves improvements over the state-of-the-art methods, which reveals that the proposed algorithm can significantly improve the accuracy of pedestrian detection in rainy scenarios.

Detection of Pedestrians in Reverse Camera Using Multimodal Convolutional Neural Networks

“…In particular, using CNN [13] has made frontal collision avoidance systems increasingly sophisticated and robust. An example of this is the research conducted by Liu et al [14], in which CNNs were employed for rain removal, and two modules were proposed: one for rain removal and another for pedestrian detection once the image was regenerated (output of the previous module). In another study [15], the authors used the YOLOv3 CNN architecture to detect pedestrians and vehicles in different scenarios.…”

Two-Stage Pedestrian Detection Model Using a New Classification Head for Domain Generalization

“…In the last decade, deep learning has enabled significant progress in a variety of applications including object detection [1,2], face recognition [3], iris recognition [4], genetic algorithms applied to CNNs [5,6], rock lithological classification [7], trademark image retrieval [8], and semantic segmentation [9], among others. Pedestrian detection is one of the key tasks in computer vision, for which several models have been developed in the past few years [10][11][12][13][14][15][16][17][18][19]. The performance has shown a steady improvement over time, especially with the boom of deep-learning-based methods, with certain benchmarks approaching human performance [20], e.g., the Caltech benchmark [21].…”

Feasibility of Using a MEMS Microphone Array for Pedestrian Detection in an Autonomous Emergency Braking System

“…For this reason, several studies are currently focused on trying to solve this problem. Some of the solutions being worked on are based on improving pedestrian detection algorithms in this type of low visibility environment [ 29 , 30 , 31 ], and others focus on using other types of systems to obtain information, such as the use of LIDAR [ 32 , 33 ] or infrared sensors [ 34 , 35 ], or on fusing images obtained from the classic RGB camera with other detection systems, such as thermal cameras [ 36 ], LIDAR [ 37 ], or an array of microphones [ 38 ].…”