Shanghai Hanshi Information Technology Co., Ltd.

Kun Ding

Sheng Huang

Shuai Yang

Yitang Zhuang

Wangpei Li

The disclosure discloses a method and a device for image stitching, the method including: performing stitching feature extraction on target images based on deep-learning neural-network model to obtain stitching features of the target images; selecting a reference image and a to-be-registered image from the target images; taking the stitching feature of the reference image under a minimum feature size and on the to-be-stitched side as a first stitching feature; calculating the similarity measure value between the first stitching feature and each of second stitching features to obtain a target stitching feature, based on the second stitching features of the to-be-registered image under the minimum feature size; repeating the above steps until the target stitching feature of the to-be-registered image under the maximum feature size is determined, and performing image stitching of the reference image and the to-be-registered image on the to-be-stitched side.

Growing community of inventors

Shanghai, China

Kun Ding