4results about How to "Avoid incomplete extraction" patented technology

The method belongs to the technical field of panoramic monitoring of ultra-high voltage converter station, and aims to solve the problem that panoramic monitoring image data is directly uploaded to the cloud, occupying a large amount of cloud resources. By adopting multi-scale convolution blocks in the deep multi-scale residual network model to construct low-order and high-order features of images of various scales, the incomplete phenomenon of image detail extraction is avoided, and the residual learning mechanism is adopted to retain low-order rough features, thereby improving the reconstruction ability of the image. Topology optimization constructs the topology structure of the heterogeneous network, and the framework combining deep reinforcement learning and Monte Carlo tree search is used to construct the network according to the pre-defined topology rules. The search result of the Monte Carlo tree strengthens the learning of the deep convolutional neural network, so as to obtain more accurate prediction in the next iteration. After the data is processed in the edge side, it is transmitted to the cloud storage, saving the cloud storage space and transmission bandwidth.

The present invention provides a method for extracting jump layers of multiple types of sound velocity profiles, belonging to the field of marine environmental survey. Based on the traditional optimal segmentation method, the present invention simplifies the sound velocity profile into a four-layer structure; according to the principle of the number of sound channel axes that meet the gradient threshold, the accurate identification of multiple jump layers is realized; taking the gradient mutation point as the starting point, the upper limit of the sound velocity jump layer is obtained through repeated iterative calculation; in the case of a single sound channel axis, the optimal segmentation algorithm is used to determine the lower limit of the jump layer; in the case of multiple sound channel axes, the optimal three-segmentation method is used to determine the upper and lower limits of the remaining jump layers. The method of the present application can effectively and accurately identify the multi-jump layer structure, and can also realize the feature extraction of multiple types of sound velocity profile jump layers, which is helpful to improve the reliability of underwater communication and the concealment performance of submarines.

The utility model relates to the technical field of agricultural products, and discloses a juice extraction device for agricultural product processing, which comprises a tank body and a hydraulic mechanism arranged at the top of the tank body and used for pressing crushed agricultural products, the pre-crushing mechanism is arranged at the upper end of the outer side of the tank body and used for dicing agricultural products, and the residue removing mechanism is arranged at the lower end of the outer side of the tank body and used for cleaning residues; an inner ring groove is formed in the lower end of the inner wall of the tank body; through cooperation of the tank body, the hydraulic mechanism, the pre-crushing mechanism and the residue removing mechanism, agricultural products put into the tank body can be crushed through the pre-crushing mechanism, and the crushed agricultural products are spirally conveyed, so that incomplete juice extraction caused by too high fiber degree of the agricultural products is avoided; meanwhile, agricultural product residues obtained after juice extraction can be scraped, the cleaning efficiency of the agricultural product residues is improved while the filter plate is prevented from being blocked, the manual cleaning cost is reduced, and the practicability is improved.

The method for super-resolution reconstruction of panoramic monitoring image of protection system of extra-high voltage converter station belongs to the technical field of power equipment detection, and solves the problems of unclearness and low resolution of the existing panoramic monitoring image, which cannot meet the demand of panoramic monitoring of inspectors; by adopting multi-scale convolution blocks to construct low-order and high-order features of images of various scales in the deep multi-scale residual network model, the phenomenon of incomplete extraction of image details is avoided, the residual learning mechanism is adopted in the network model to retain low-order rough features, reduce the training difficulty, promote the reuse of features, and thus improve the reconstruction capability of the image; the reconstructed image has better structural similarity and peak signal-to-noise ratio performance; the standard data set and the panoramic monitoring image data set of the extra-high voltage converter station are used in sequence for image super-resolution reconstruction and target recognition experiment, and the experimental results show that the high-resolution image reconstructed by the method can meet the demand of panoramic monitoring of inspectors.