Method for accurately positioning candidate bounding boxes in target segmentation network

Abstract

The invention discloses a method for accurately positioning candidate bounding boxes in a target segmentation network, and the method comprises the steps: 1, constructing a lightweight convolutional neural network model, and designing network parameters; 2, solving an intersection-to-parallel ratio IoU of an IoU layer in the network model; 3, solving confidence scores of candidate bounding boxes in the network model; and 4, removing the redundancy of the candidate bounding box according to the setting of the intersection-parallel ratio and the confidence score, and finally enabling the confidence score of the output candidate bounding box to be highest and closest to a real calibration box. According to the method, a balance position can be found between target missing detection and errordetection, the candidate bounding box is closer to a real calibration box, and meanwhile, the network structure and the confidence coefficient solving method designed by the invention have certain generalization ability and can be applied to a relatively complex scene.

Description

technical field [0001] The invention relates to a method for precise positioning of candidate bounding boxes in a target segmentation network. [0002] technical background [0003] When applying machine learning to segment the target object in the image, it is first necessary to scale the input original image, and then input the fixed-size image into the pre-set convolutional neural network for feature extraction, and then in the feature map Anchor points are performed on the object, and then the region of interest is extracted to determine the candidate bounding boxes of the foreground (target object) and the background. [0004] After the above-mentioned process, the four-dimensional position coordinates of the candidate bounding box (center position coordinates and its width and height) are often obtained. There is a great deviation from the real frame where the pre-calibrated target is located, and it also leads to two problems. . First, if the overlap rate between the...

AI Technical Summary

Problems solved by technology

[0004] After the above process, the four-dimensional position coordinates of the candidate bounding box (center position coordinates and its width and height) often obtained have a great deviation from the real frame where the pre-calibrated target is located, and it also leads to two problems.

First, if the overlap rate between the obtained candidate bounding box and the real calibration box is greater than the predetermined threshold, it will cause the candidate bounding box to be eliminated, and it is very likely that some targets will be missed.

Second, it is often difficult to pre-set the overlap rate threshold between the candidate bounding box and the real calibration box. If the setting is too small, some targets may be missed. If the threshold is set too large, it is easy to cause the same target. The case of multiple detections or false detections

Images