Target detection method and system based on fusion of different-scale receptive field feature layers, and medium

Abstract

The invention provides a target detection method and system based on different-scale receptive field feature layer fusion and a medium. The method comprises the following steps: a data volume increasing step: carrying out incremental processing on a training data set with a label, increasing the data volume of the training data set, adjusting the training image size of the training data to be thesame as the model input size, and obtaining the training data set after data increase; and a target detection network model building step: taking the classic network model as the network basis of thetarget detector, and replacing the transverse connection in the feature pyramid network FPN with the dense connection to obtain a dense connection FPN target detection network model. The defect that an existing target detection model only uses feature information in part of feature layers to detect a target object is overcome; the feature layers of a plurality of different receptive fields are fused through FPN dense connection, so that feature information required for object detection in a plurality of scale ranges can be obtained, and the feature extraction capability and the target detection performance of the target detector are improved.

Description

technical field [0001] The invention relates to the field of intelligent detection and recognition of target objects in images, in particular to a target detection method, system and medium for fusion of feature layers of receptive fields of different scales. In particular, it relates to a target detection method based on the fusion of feature information in different feature layers based on deep learning Background technique [0002] Object Detection is an important basic research field in computer vision. Its main work is to locate the object of interest (ROI) in the image (Localization) and classify the category of the object ROI (Classification). Before the emergence of the convolutional neural network model (CNN), the main research method of target detection was to manually extract the feature information required for target object detection in images, while the deep learning-based target detector (CNN-based Object Detector) relies on Its excellent feature extraction a...

AI Technical Summary

Problems solved by technology

[0004] Although FPN solves the problem of spatial information loss to a certain extent, the detection effect of extreme-scale objects is still not ideal. The study found that the feature information required for object detection at a certain scale is not only distributed on the feature layer corresponding to its scale, but also There is a large amount of relevant information distributed on other feature layers, and only obtaining the feature information contained in a single-layer feature layer cannot meet the needs of target detection

Images