Hangzhou Hikvision Digital Technology Co., Ltd.

Yi Niu

Binghua Luo

Zhanzhan Cheng

Shiliang Pu

Gang Zheng

Yunlu Xu

Zuozhou Pan

An image processing method and apparatus for target recognition is disclosed. The method includes: inputting N frames of images into a quality evaluation network model; determining, with the quality evaluation network model, a feature vector of each of the N frames of images according to an attention weight of a preset sub-region image and M quality evaluation parameters; determining quality evaluation values of the N frames of images according to the feature vectors of the N frames of images; determining a target image or a target vector for target recognition according to the quality evaluation values of the N frames of images. The feature vector of each frame of image is determined based on the attention weight of the sub-region image and M quality evaluation parameters.

Image processing method and apparatus for target recognition

An image object recognition method, apparatus, and computer device are provided. The image object recognition method includes: performing feature extraction in the direction of a horizontal angle of view and in the direction of a vertical angle of view of an image respectively, to extract a lateral feature sequence and a longitudinal feature sequence of the image; fusing the lateral feature sequence and the longitudinal feature sequence to obtain a fused feature; activating the fused feature by using a preset activation function to obtain an image feature; and recognizing an object in the image by decoding the image feature. This solution can improve the efficiency of the object recognition.

Lateral and longitudinal feature based image object recognition method, computer device, and non-transitory computer readable storage medium

An object recognition method and apparatus for a deformed image are provided. The method includes: inputting an image into a preset localization network to obtain a plurality of localization parameters for the image, wherein the preset localization network comprises a preset number of convolutional layers, and wherein the plurality of localization parameters are obtained by regressing image features in a feature map that is generated from a convolution operation on the image; performing a spatial transformation on the image based on the plurality of localization parameters to obtain a corrected image; and inputting the corrected image into a preset recognition network to obtain an object classification result for the image. In the process of the neural network based object recognition, the embodiment of the present application first transforms the deformed image that has deformation, and then performs the object recognition on the transformed image.

Target recognition method and apparatus for a deformed image

Embodiments of the present disclosure provide a method and apparatus for detecting a license plate. The method comprises: obtaining, according to pixel values of pixels in an image to be detected, a candidate license plate area Min the image to be detected (S); calculating an aspect ratio of the candidate license plate area Mand determining if the aspect ratio is greater than a first predefined threshold (S); determining, if the aspect ratio is greater than the first predefined threshold, a new candidate license plate area Mfrom the candidate license plate area Maccording to a predefined machine learning-based regression algorithm (S); determining, according to a first predefined classification model, whether the candidate license plate area Mis a license plate area (S); determining, if the candidate license plate area Mis a license plate area, the candidate license plate area Mas a license plate area, and generating a detection result based on the candidate license plate area M(S). The application of the solution improves the accuracy of license plate detection.

Growing community of inventors

Hangzhou, China

Yi Niu