SR-Det:面向工业场景下细长和旋转目标的鲁棒检测

SR-Det:Towards Robust Detection of Slender and Rotated Objects in Industrial Scene

目标检测广泛应用于工业领域,譬如缺陷检测。然而,在检测过程中依然存在任意旋转和大宽高比问题。一是水平锚框方法难以准确地定位物体;二是卷积神经网络(Convolutional Neural Networks,CNNs)在提取特征时表现不佳;三是普通的损失函数对细长的目标不敏感。针对上述问题,本文研究了SR-Det (Slender and Rotated Detecto)模型,包含以下3个部分。首先是旋转区域校准(Rotated Region Calibration,RRC)模块。该算法以不同大小和宽高比的水平提议作为输入,以相应的旋转提议作为输出。然后是旋转角度提议对齐模块(Rotated Angle Proposal Align,RAP-Align)来保证特征信息的质量。最后是基于交并比(Intersection Over Union,IoU)策略的R-IoU函数(Rotated Intersection Over Union)以指导模型最大化预测框和GT (Ground Truth)框之间的重叠面积。实验证明,本文提出的方法在金属罐数据集和幕墙数据集上取得了最优的效果,证明了该方法的有效性。

Though object detection has been widely used in the industrial scene,it still faces the detection problems of crack defects with slender and rotated characteristics.On the one hand,traditional horizontal anchor methods are usually hard to precisely locate the object.On the other hand,CNNs(Convolutional Neural Networks)perform poorly in terms of feature extraction from rotated objects.In addition,normal loss functions are insensitive to slender objects.To address these,this paper proposes a Slender and Rotated Detector(SR-Det)for robust slender and rotated object detection.Specifically,the Rotated Region Calibration(RRC)is designed,which takes horizontal proposals with different scales and aspect ratios as inputs and outputs the corresponding rotation proposals.Then,the Rotated Angle Proposal Align(RAP-Align)is presented to guarantee the quality of extracted feature information.Finally,the Rotated intersection over union(R-IoU)based on Intersection Over Union(IoU)strategy is proposed for guiding the model to maximize the area between predicted box and Ground Truth box.The experiments on metal cans and curtain walls datasets have shown that the method proposed achieves state-of-the-art performance,demonstrating the effectiveness of the proposed algorithm.