Method, system and equipment for dense target detection in any direction based on feature focusing

Abstract

The invention discloses a method, system and equipment for dense target detection in any direction based on feature focusing. The method includes four steps. One is to focus on the target features in the training set, complete data preprocessing, and generate a new training set; It is to build a target detection network and train the feature extraction ability of the model; the third is to extract the features of the target in the picture to be detected from different directions for detection; the fourth is to synthesize the detection results of each direction. The model constructed by the present invention fully focuses on the target features and realizes the accurate extraction of features. The experimental results show that the detection model generated by this method not only improves the accuracy of the target detection in the densely arranged direction, but also reduces the missed detection. Rate.

Description

technical field [0001] The present invention relates to a method, system and equipment for dense target detection in any direction, in particular to a method, system and equipment for dense target detection in any direction based on feature focusing. Background technique [0002] There are various types of information. In particular, information in the form of images is particularly important. Compared with information such as text and audio, images are more intuitive and contain more information. Fully extracting image information is an important direction of future information processing. The traditional way of extracting image information is mainly manual interpretation, which is inefficient. With the development of science and technology, image acquisition capabilities are getting stronger and stronger, and the number of images is increasing exponentially. Artificial methods can no longer adapt to the development of the times and the actual situation, and it is urgent to...

AI Technical Summary

Problems solved by technology

[0004] The existing mainstream detection model algorithms have poor detection results when dealing with densely arranged targets with arbitrary directions, especially when facing some targets with large aspect ratios.

In real life, the shape of some targets (such as airplanes) tends to be square, and the distribution is relatively scattered. When detecting the square or fine-tuning the square, the position of the target can be well located, so that the target can be successfully detected, but for For other targets (such as ships), the aspect ratio of the target is often very large, and the arrangement is dense and the direction is arbitrary. For these targets, it is difficult for the mainstream detection algorithm to accurately locate the specific position of the target, and the given detection frame is often Adjacent targets will be covered, the overlap between the detection frames is serious, and there are many missed detections, so the detection effect is not good

Images