Fast hash vehicle retrieval method based on multi-task deep learning

Abstract

The invention provides a fast hash vehicle retrieval method based on multi-task deep learning. The fast hash vehicle retrieval method includes a multi-task deep convolutional neural network used for deep learning and training recognition, a segmented compact hash code and instance feature fusion method for improving the retrieval accuracy and the practicality of the retrieval method, a local sensitive hash reordering algorithm for improving the retrieval performance and a cross-modal retrieval method for improving the robustness and accuracy of a retrieval engine. In the fast hash vehicle retrieval method, firstly, a method for segmented learning of hash codes through a multi-task deep convolutional network is proposed, image semantics and image representation are combined, the connectionbetween related tasks is used for improving the retrieval accuracy and refining image features, and at the same time, minimizing image coding is used for making learned vehicle features more robust; secondly, a feature pyramid network is used for extracting the instance features of vehicle images; then, a local sensitive hash reordering method is used for retrieving the extracted features; and finally, a cross-modal assisted vehicle retrieval method is used for the special case in which target images of inquired vehicles cannot be obtained.

Description

Technical field [0001] The invention relates to the application of artificial intelligence, digital image processing, convolutional neural network and computer vision in the field of public safety, and belongs to the field of intelligent transportation. Background technique [0002] Today, with the rapid development of smart cities and intelligent transportation, the demand for vehicle identification and vehicle retrieval in large-scale image monitoring and video databases in public safety systems has increased dramatically. [0003] In the prior art, the vehicle retrieval method mainly extracts the license plate information of the target vehicle. Then, the motor vehicle to be retrieved is retrieved based on the license plate information. The general specific method is to identify the license plate number of the vehicle from the surveillance image, and then identify the motor vehicle with the license plate number in other surveillance images. Although this method of searching onl...

AI Technical Summary

Problems solved by technology

[0015] Aiming at how to efficiently utilize massive video data generated in the field of public security and improve the efficiency of vehicle retrieval in the era of big data, the present invention proposes a fast hash retrieval method based on multi-task deep learning, which effectively utilizes the gap between detection and recognition tasks. Relevance and diversity of bayonet vehicle basic information to achieve the purpose of real-time retrieval; finally provide a multi-task deep learning fast hash vehicle retrieval method with high retrieval accuracy and good robustness

Images