Integrated detection and fine-grained recognition method for ship targets in remote sensing images

Abstract

The embodiment of the invention discloses a method for integrated detection and fine-grained identification of a remote sensing image ship target. The method includes: acquiring a training sample set; determining a feature extraction shared model integrated in a remote sensing image processing model, a location detection branch model for outputting the location information of the target object, and a type recognition branch model for outputting the type information of the target object ; Adopt the shared image features extracted from the training samples by the feature extraction shared model to alternately train the position detection branch model and the type recognition branch model, so as to complete the training of the remote sensing image processing model when the preset convergence conditions are met. . By adopting the solution of the present application, it is possible to avoid repeated extraction of the shared part of the image features during training or use of the position detection branch model and the type recognition branch model, avoid additionally increasing the task load of this part of the extraction operation, and not only greatly reduce the model training. amount of tasks.

Description

technical field [0001] Embodiments of the present invention relate to the technical field of remote sensing image processing, and in particular, to a method for integrated detection and fine-grained identification of ships in remote sensing images. Background technique [0002] Ship target detection and recognition based on remote sensing images is an important means of ocean monitoring, and it has a pivotal research significance in various fields. [0003] Currently, in the field of computer vision, current deep learning-based methods are becoming the mainstream technology for object detection and object recognition tasks. However, each target detection task and target recognition task are completed independently, and the detection and recognition of ship targets cannot be completed at the same time. More importantly, the target detection task and target recognition task will have something in common, whether it is model training. Still using the model, performing tasks in...

AI Technical Summary

Problems solved by technology

However, each target detection task and target recognition task are completed independently, and the detection and recognition of ship targets cannot be completed at the same time. More importantly, the target detection task and the target recognition task will have something in common. Still using the model, performing tasks independently will increase the amount of additional tasks, and will reduce the efficiency of target detection and recognition when using the model

Images