Satellite image segmentation method based on residual network and U-Net segmentation network

Abstract

The invention discloses a satellite image segmentation method based on residual network and U-Net segmentation network. The satellite image segmentation method comprises the following steps: constructing a residual network ResNet 34; constructing U-Net segmentation network; constructing a training sample set; training the residual network ResNet 34; training the U-Net segmentation network; inputting the satellite image to be segmented into the residual network ResNet 34 for binary classification, and judging that a ship target is included; using the U-Net segmentation network to perform binarysegmentation on the positive samples in the classification result; for the negative samples in the classification result, directly outputting a single-value mask graph. According to the method, the satellite image is subjected to binary classification by using the residual network ResNet 34, and the U-Net segmentation network is used to only segment the positive samples in the classification result, and an SE-ResNet module is embedded in the U-Net segmentation for extracting finer segmentation masks and the satellite image segmentation method is high in real-time performance and segmentationprecision.

Description

technical field [0001] The invention belongs to the technical field of image processing, and further relates to a satellite image segmentation method based on a residual network and a U-Net segmentation network in the technical field of image segmentation. The invention can be used to detect the ship target from the high-resolution satellite image and segment the area where the ship is located from the image. Background technique [0002] With the continuous development of today's society, maritime transportation safety has become a hot topic of concern. With the growing demand for shipping, more ships at sea have increased the possibility of illegal transportation at sea, such as illegal fishing, piracy, and illegal cargo transportation, which has caused great difficulties for maritime supervision. Ship detection and segmentation in satellite images can help regulatory authorities monitor ships at sea in real time and maintain maritime transportation safety. However, due ...

AI Technical Summary

Problems solved by technology

The disadvantage of this method is that only grayscale histogram features are extracted, which are greatly affected by illumination and noise, and only contain underlying grayscale and texture information, which has poor robustness, resulting in poor segmentation accuracy. high

The disadvantage of this method is that all satellite images are directly input into the network for segmentation and classification. These images often contain a large number of negative samples, and a large number of negative sample images are input into the network for segmentation. The real-time performance is poor, resulting in a waste of space resources. ;In addition, the use of binary cross-entropy and Jaccard joint loss function can not train the segmentation network very well, because there are few target areas and many background areas, binary cross-entropy and Jaccard joint loss function are easily dominated by a large number of background areas, resulting in Divided networks are not easy to converge

Images