Pedestrian detection method based on multi-scale self-attention feature fusion

Abstract

The invention discloses a pedestrian detection method based on multi-scale self-attention feature fusion, and relates to the field of artificial intelligence. The method comprises the following steps: (1) acquiring data of a pedestrian detection image; (2) designing the size of a pedestrian in the detection image; (3) dividing positive and negative samples of a pedestrian data set; and (4) building a pedestrian detection model. According to the method, the Faster R-CNN is adopted for building a pedestrian detection framework, and a multi-scale feature fusion network model is provided, so that more and more effective feature information can be extracted and overfitting can be avoided; a GPU with good performance is used for training, so that the training speed is greatly increased; the receptive field is expanded, so that small targets can be detected and the resolution ratio is not reduced; and the method is very suitable for accurate and rapid detection of pedestrians.

Description

Technical field: [0001] The invention relates to a pedestrian detection method for multi-scale self-attention feature fusion, which belongs to the field of artificial intelligence. Background technique: [0002] Pedestrian detection, as a field of great research value in computer vision, is widely used in unmanned driving, human behavior analysis, intelligent transportation, intelligent video surveillance and other fields; it is a vehicle assisted driving, intelligent video surveillance and human behavior analysis The first step in the application has also been applied in emerging fields such as aerial photography and victim rescue in recent years; due to the highly arbitrary posture of the human body, complex and changeable shapes, and problems such as attachment and occlusion, it can accurately detect in various scenarios The theory and technology of pedestrians still need to be further explored and researched; the process of pedestrian detection is: to detect the input pi...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the low detection efficiency and poor detection effect of the traditional pedestrian target detection algorithm, a pedestrian detection algorithm based on multi-scale self-attention feature fusion is proposed.

Images