Small-target human body detection method based on balanced sampling and nonlinear feature fusion

Abstract

The invention provides a small-target human body detection method based on balanced sampling and nonlinear feature fusion, and the method comprises the steps: carrying out the fusion and enhancement of the features of all scales in a training process, and enabling the scale where a small-target human body is located to obtain sufficient human body semantic information; using an instance balance sampling strategy for positive samples during subsequent loss calculation, it is guaranteed that human bodies of different scales can be trained in the same way, an intersection-parallel ratio balance sampling strategy is used for negative samples, it is guaranteed that enough samples are difficult to participate in training, and finally the classification capacity of the network is improved; and aknowledge distillation training method is adopted in a training strategy, so that the model size is compressed while the precision is ensured, and the reasoning speed is increased.

Description

technical field [0001] The invention relates to the field of target detection, in particular to a small target human body detection method based on balanced sampling and nonlinear feature fusion. Background technique [0002] Object detection is an important research field in computer vision, which includes two processes: classification and localization. The classification process outputs the category of the target, and the positioning process produces the coordinate information of the target. Human detection is an important branch of target detection. The goal of human detection is to detect whether there is a human target in the image scene and give the position of the target. It has a wide range of applications in the fields of automatic driving, video surveillance, and mobile robots. [0003] At present, target detection methods based on deep learning can be roughly divided into two categories: single-stage and two-stage: single-stage detection methods regard target det...

AI Technical Summary

Problems solved by technology

[0004] However, when detecting human objects, due to the large number of small objects in human detection, the number of negative samples that are simple and easy to classify is too large, resulting in poor classification ability of the network; and the existing feature fusion method only fuses high-level feature information. In the lower layer, the fusion of feature information of each layer is insufficient, resulting in the lack of semantic information of the human body for small target objects, which affects the detection effect

Images