This paper presents an edge detection algorithm based on singular value eigenvector and gradient operator. In the proposed algorithm, the singular values of image blocks are first calculated, and the Sobel gradient template is extended to eight other directions. Then the gradient values of image pixels are determined according to the stability of the singular values of image blocks. The determination of gradient threshold is considered from both global and local aspects. After calculating the global and local gradient thresholds of the original image, the gradient threshold of the whole image is determined by weighting function. Then the edge pixels of the image are filtered according to the gradient threshold, and the edge information image of the original image is obtained. The experimental data show that the proposed algorithm can resist a certain degree of noise interference, and the accuracy and efficiency of edge extraction are better than other similar algorithms.

Single-frame image super-resolution (SISR) technology in remote sensing is improving fast from a performance point of view. Deep learning methods have been widely used in SISR to improve the details of rebuilt images and speed up network training. However, these supervised techniques usually tend to overfit quickly due to the models’ complexity and the lack of training data. In this paper, an Improved Deep Recursive Residual Network (IDRRN) super-resolution model is proposed to decrease the difficulty of network training. The deep recursive structure is configured to control the model parameter number while increasing the network depth. At the same time, the short-path recursive connections are used to alleviate the gradient disappearance and enhance the feature propagation. Comprehensive experiments show that IDRRN has a better improvement in both quantitation and visual perception.

Current mainstream super-resolution algorithms based on deep learning use a deep convolution neural network (CNN) framework to realize end-to-end learning from low-resolution (LR) image to high-resolution (HR) images, and have achieved good image restoration effects. However, as the number of layers in the network is increased, better results are not necessarily obtained, and there will be problems such as slow training convergence, mismatched sample blocks, and unstable image restoration results. We propose a preclassified deep-learning algorithm (MGEP-SRCNN) using Multilabel Gene Expression Programming (MGEP), which screens out a sample sub-bank with high relevance to the target image before image block extraction, preclassifies samples in a multilabel framework, and then performs nonlinear mapping and image reconstruction. The algorithm is verified through standard images, and better objective image quality is obtained. The restoration effect under different magnification conditions is also better.

With the rapid expansion of data volume, traditional data storage methods have been unable to meet the practical application requirements of blockchain cloud storage. Aiming for the cloud storage problem of blockchain, a new storage access method for predicting dynamic file load is proposed. By predicting the load status of cloud storage files in advance, the load of each blockchain data node at the next moment is first estimated. A hierarchical genetic algorithm is used to construct the connection weights between the hidden layer and the output layer, which makes the data network converge faster and more accurate, thereby effectively predicting the node load. In addition, based on the file allocation, an evaluation analysis model is constructed to obtain the time response capability of each file during the allocation process. The node’s periodic load prediction value is used to calculate the corresponding weight of the node and it is continuously updated, retaining the advantages of the static weighted polling algorithm. Combined with the genetic algorithm to help predict the file assignment access strategy of the later load of each node, it can meet the system requirements under complex load conditions and provide a reasonable and effective cloud storage method. The experimental evaluation of the proposed new strategy and new algorithm verifies that the new storage method has a faster response time, more balanced load, and greatly reduced energy consumption.