An improved Mask R-CNN image instance segmentation method for identifying defects of power equipment

Abstract

The invention discloses an improved Mask R-CNN image instance segmentation method for identifying defects of power equipment. The method comprises the steps of constructing a convolutional neural network; reading the power equipment defect pre-processed picture, inputting the power equipment defect pre-processed picture into a convolutional neural network, and performing feature extraction on thepower equipment defect pre-processed picture by the convolutional neural network to obtain a feature-containing region; refining the region containing the characteristics through an RPN network; and in the refined region, carrying out RoIAgign operation through a bilinear difference method to process the refined region, generating a feature map of a fixed size for each region of interest, and obtaining power equipment defect instance segmentation through the category, the coordinate information and the mask information. By improving the Mask R-CNN image instance segmentation method, the basiceffect of Mask R-CNN segmentation is reserved, the bilinear interpolation speed in the RoIAgign process is increased, meanwhile, the mapping process fully and uniformly utilizes power equipment defects to preprocess all pixels of an image, and the segmentation effect is more obvious.

Description

technical field [0001] This application relates to the technical field of image target detection and segmentation, and in particular to an improved Mask R-CNN image instance segmentation method for identifying defects in electrical equipment. Background technique [0002] With the rapid development of the power grid, electrification is also developing faster and faster. In order to ensure the normal operation of the power equipment, it is not feasible to shut down the power equipment during operation and conduct relevant inspections on it. At the same time, some hidden dangers and defects of power equipment are often latent, so they cannot be discovered in time, let alone eliminated in time. [0003] When the occlusion phenomenon occurs between the targets of the power equipment, if the nearest neighbor interpolation method (Nearest Neighbor Interpolation) is simply used, although the calculation amount is small and the speed is fast, the spatial symmetry (Alignment) is dest...

AI Technical Summary

Problems solved by technology

Since MaskR-CNN is segmented pixel by pixel, when the bilinear interpolation algorithm is used, the accuracy of image segmentation will be improved, but at the same time, it will lead to disadvantages such as reduced calculation speed and slower segmentation speed, so the practicability of the defect detection system for electrical equipment is limited. limit

Images