Detection method for picture small target

Abstract

The invention discloses a detection method for a picture small target. The method includes using a to-be-detected picture as the input of a target detection network; obtaining six feature maps with different sizes from the to-be-detected picture, performing feature fusion on the bottom-layer feature map of the pyramid and the high-layer feature map of the pyramid in the six feature maps with different sizes by adopting a bilinear interpolation method to obtain six new feature maps with different sizes; considering the relationship among the scale feature maps; the feature map participating inprediction has richer detail information and semantic information; training the target detection network by using default setting of a pre-training model, using a random gradient descent algorithm tocarry out optimization training on the loss function expression, so rapid convergence of the model is facilitated, the trained target detection network is utilized to carry out testing, to-be-detectedimage characteristics are extracted, detection precision and detection speed of the picture small target can be effectively improved, and detection precision of the small target can be effectively improved under the condition of completely satisfying real-time performance.

Description

technical field [0001] The invention belongs to the field of target detection, and in particular relates to a detection method for a small target in a picture. Background technique [0002] Object detection is one of the core issues in the field of computer vision and an important research direction in computer vision. With the rapid development of computer vision, object detection is widely used in intelligent transportation, medical image diagnosis, image retrieval, and military applications. [0003] Traditional target detection algorithms can generally be divided into three stages: region selection, feature extraction, and classifier classification. Region selection mostly adopts the sliding window strategy to traverse the entire image, and the windows need to be set with different scales and different aspect ratios. Although this exhaustive strategy includes all possible positions of the target, it takes too long, has high window redundancy and high cost. Due to fact...

AI Technical Summary

Problems solved by technology

The traditional target detection algorithm has great shortcomings: 1. The region selection strategy based on the sliding window is not targeted, the time complexity is high, and the window is redundant; 2. The manually designed features are not very good for the change of diversity. robustness

However, since the SSD detection model only has the underlying conv4_3 for detecting small targets, it does not have enough semantic information, and does not take into account the relationship between feature maps of different sizes, resulting in poor detection performance for small targets, so it still needs to be explored A New Way to Improve the Detection Accuracy of Small Objects

Images