Vehicle-Damage-Detection Segmentation Algorithm Based on Improved Mask RCNN

Qinghui Zhang,Shanfeng Bian Bian,Xianing Chang

doi:10.1109/access.2020.2964055

Abstract

Traffic congestion due to vehicular accidents seriously affects normal travel, and accurate and effective mitigating measures and methods must be studied. To resolve traffic accident compensation problems quickly, a vehicle-damage-detection segmentation algorithm based on transfer learning and an improved mask regional convolutional neural network (Mask RCNN) is proposed in this paper. The experiment first collects car damage pictures for preprocessing and uses Labelme to make data set labels, which are divided into training sets and test sets. The residual network (ResNet) is optimized, and feature extraction is performed in combination with Feature Pyramid Network (FPN). Then, the proportion and threshold of the Anchor in the region proposal network (RPN) are adjusted. The spatial information of the feature map is preserved by bilinear interpolation in ROIAlign, and different weights are introduced in the loss function for different-scale targets. Finally, the results of self-made dedicated dataset training and testing show that the improved Mask RCNN has better Average Precision (AP) value, detection accuracy and masking accuracy, and improves the efficiency of solving traffic accident compensation problems.

Highlights

Object detection is one of the main research contents of computer vision
The positioning ability of the detection frame is limited, and when the feature is extracted, as the number of convolution layers increases, gradient disappearance or gradient explosion often occurs. He Kaiming et al proposed a residual network (ResNet) [5] [25], which helps the model to converge by using the residual module, accelerates the training of the neural network, and combines with the target detection model Mask RCNN[6] [26] [27] to realize object detection and segmentation, greatly improving the accuracy of the model detection
EXPERIMENTAL RESULTS AND ANALYSIS To reduce the number of steps in making dataset labels and to improving the detection accuracy of car-damage images, transfer learning and Mask RCNN are used in this paper to process and detect images showing damage

Summary

INTRODUCTION

Object detection is one of the main research contents of computer vision. It is to determine the category and location information of the object of interest in the image on the instance level. FPN [21] uses a top-down hierarchy with lateral connections, from single-scale input to building a network feature pyramid, which solves the multi-scale problem of extracting target objects in images This structure has strong robustness and adaptability, and requires fewer parameters. When the pixel-level segmentation is directly performed, the image target object cannot be accurately positioned, so the Mask RCNN is improved on the basis of Faster RCNN, and the Rol Pooling layer is changed into the interest-region alignment layer (RoIAlign). The values of the four positions are calculated by bi-linear interpolation, and the maximum pooling or average pooling operation performed to obtain the feature map of 2 × 2 size

IMPROVEMENT OF LOSS FUNCTION The multitasking loss function of Mask RCNN is

Ncls i

CONCLUSION