Beijing University of Posts and Telecommunications

Huiyuan Fu

Huadong Ma

Wu Liu

Haitao Zhang

Zheng Zhang

Yu Cao

Cheng Zhang

Yifan Zhang

Xinchen Liu

Dengkui Zhang

Embodiments of the present invention provide an end-to-end lightweight method and apparatus for license plate recognition. The method comprises: obtaining an image to be recognized; obtaining a number of a license plate in the image to be recognized and position coordinates of the license plate in the image to be recognized on the basis of the image to be recognized and a pre-trained target license plate recognition model, wherein the target license plate recognition model comprises a target feature extraction network, a target region candidate localization network, a target super-resolution generation network and a target recurrent neural network. Because in this solution, once an image to be recognized is input into the target license plate recognition model, the target license plate recognition model can output the license plate number and position coordinates of the license plate in the image to be recognized, one realizes an end-to-end model. The model has relatively strong robustness, and it can detect and recognize pictures taken under different camera angles. Moreover, computation variables such as image features can be reused without repeated computations, the model takes up less RAM and the speed of license plate recognition is greatly improved.

End-to-end lightweight method and apparatus for license plate recognition

The present application discloses a vehicle searching method and device, which can perform the steps of: calculating an appearance similarity distance between a first image of a target vehicle and several second images containing the searched vehicle; selecting several images from the several second images as several third images; obtaining corresponding license plate features of license plate areas in the first image and each of the third images with a preset Siamese neural network model; calculating a license plate feature similarity distance between the first image and each of the third images according to license plate feature; calculating a visual similarity distance between the first image and each of the third images according to the appearance similarity distance and the license plate feature similarity distance; obtaining a the first search result of the target vehicle by arranging the several third images in an ascending order of the visual similarity distances. The solution provided by the present application is not limited by application scenes, and it also improves vehicle searching speed and accuracy while reducing requirements of hardware such as cameras that collect images of a vehicle and auxiliary devices.

Progressive vehicle searching method and device

The invention relates to a weather recognition method and device based on image information detection, including: obtaining an image extracting multiple first image features of the image with respect to each preset type of weather using a number of first preset algorithms preset correspondingly for different preset types of weather; inputting the multiple first image features to a preset multi-kernel classifier, the multi-kernel classifier performing classification according to the image features to identify the weather in which the image was taken. The multi-kernel classifier is realized by: selecting a first preset number of image samples for each of the preset types of weather; for the image samples of this type of weather, extracting the first image features of each image sample according to the first preset algorithm corresponding to this preset type of weather; and performing machine learning for the first image features according to a preset multi-kernel learning algorithm.

Weather recognition method and device based on image information detection

Disclose is a gait recognition method, firstly, extracting an initial gait feature of a gait video of a person to be recognized; secondly obtaining a corresponding optimized gait feature according to a trained sub neural network and the initial gait feature; then determining corresponding degrees of similarity according to the optimized gait feature of the person to be recognized and the optimized gait feature of each known person in a matching library, and determining information of the person to be recognized according to information of the known person in the matching library corresponding to the optimized gait feature which has the highest degree of similarity with the optimized gait feature of the person to be recognized.

Growing community of inventors

Beijing, China

Huiyuan Fu