Parallel method of object detection and semantic segmentation based on end-to-end depth learning

Abstract

The invention provides a parallel method of target detection and semantic segmentation based on end-to-end depth learning, which obtains a model compose of a target detection neural network Darknet-19, a target segmentation full convolution neural network FCN by training a massive labeled target detection frame and a pixel-level target segmentation image. , and tasks of parallel target detection and target segmentation are successfully realized. That method can effectively extract image features, realizes real-time image processing functions on the basi of ensuring target detection and targetsegmentation, and has a wide application prospect.

Description

technical field [0001] The invention relates to the field of computer vision of artificial intelligence, in particular to a parallel method of target detection and semantic segmentation based on end-to-end deep learning. Background technique [0002] In the field of deep learning target detection, two problems are mainly solved. One is the classification and positioning of multiple targets in the image. Its development process is divided into three stages. One is the traditional target detection method, and the other is Regions with CNNfeatures (R-CNN) is a target detection framework that combines region proposals and CNN classification, such as: Fast R-CNN, Faster R-CNN, and the third is You Only Look Once (YOLO) as a representative of the target The end-to-end (End to End) target detection framework that detects regression problems, such as: SSD; the traditional method has a sliding window-based region selection strategy that lacks pertinence and hand-designed features are...

AI Technical Summary

Problems solved by technology

[0004] For the existing deep learning target detection, the region selection strategy based on the sliding window lacks pertinence and hand-designed features are not robust to diversity changes. At the same time, the semantic segmentation field of deep learning performs pixel segmentation on multiple types of target objects in the image. The division of levels, the problem of insufficient division of the boundary information of the target object, the present invention proposes a natural language sentiment analysis method based on a deep network, the technical solution adopted by the present invention is:

Images