To handle the issues of small sample size and multi-target detection in the hierarchical detection of diabetic retinopathy, a weakly supervised target detection network based on improved GoogLeNet was proposed. Firstly, the GoogLeNet network was improved, the last fully-connected layer of the network was removed and the position information of the detection target was retained. A global max pooling layer was added, and the sigmoid cross entropy was used as the objective function of training to obtain the feature map with multiple feature position information. Secondly, based on the weak supervision method, only the category label was used to train the network. Thirdly, a connected region algorithm was designed to calculate the boundary coordinate set of feature connected regions. Finally, the boundary box was used to locate the lesion in the image to be tested. Experimental results show that under the small sample condition, the accuracy of the improved model reaches 94%, which is improved by 10% compared with SSD (Single Shot mltibox Detector) algorithm. The improved model realizes end-to-end lesion recognition under small sample condition, and the high accuracy of the model ensures its application value in fundus screening.

In the processing of registration for multi-temporal high resolution remote sensing images, the phenomena of surface features change and relative parallax displacement caused by differences in acquisition conditions degrades the accuracy of registration. To resolve the aforementioned issue, a registration algorithm for multi-temporal high resolution remote sensing images based on abnormal region sensing was proposed, which consists of coarse and fine registration. The algorithm of Scale-Invariant Feature Transform (SIFT) has a better performance on scale space, the feature points from different scale space indicates the various size of spot. The high scale space points represent the objects which have a stable condition, the coarse registration can be executed depending on those points. For the fine registration, intensity correlation measurement and spatial constraint were used to decide the regions which were used to extract the efficacious points from low scale space, the areas for searching matching points were limited as well. Finally, the accuracy of the proposed method was evaluated from subjective and objective aspects. Experimental results demonstrate that the proposed method can effectively restrain the influence of abnormal region and improve registration accuracy.