Abstract: The application discloses a method for optimizing a lithium-potassium anticline structure target area, which includes the following steps: obtaining data of a historical lithium-potassium anticline structure area, and classifying the data of the historical lithium-potassium anticline structure area to generate classified data; carrying out a parameter assignment on the classified data to obtain a parameter data set; constructing a neural network model, inputting the parameter data set into the neural network model for a training, and obtaining a target area optimal neural network model; and based on the target area optimal neural network model, carrying out a target area optimization in a deep lithium-potassium anticline structure area, and obtaining an optimal result.

Abstract: The present disclosure provides a potassium dihydrogen phosphate (KDP) frequency-doubling crystal structure, including a crystal fixing structure, a KDP crystal, a ? polarization tube, and a ? polarization tube; where a ?-polarized laser with a central wavelength of 800 nm enters the KDP crystal through the ? polarization tube, and the KDP crystal converts the ?-polarized laser with a central wavelength of 800 nm into a ?-polarized laser with a central wavelength of 400 nm by frequency doubling and then outputs the ?-polarized laser from the ? polarization tube. Close end bevels of the ? polarization tube and the ? polarization tube are designed into a Brewster angle. The KDP crystal is placed in a vacuum-sealed cavity formed by the ? polarization tube, the ? polarization tube, and the crystal fixing structure, which eliminates a complicated process of crystal coating and prevents the KDP crystal from deliquescing in air.

Abstract: This application relates to a ternary precursor for making a material for positive electrodes in lithium batteries. In the ternary precursor, primary particles or whiskers of the ternary precursor are distributed in a radial direction. A deformation stacking fault probability fD of the ternary precursor is ?2.5%. This application further relates to a preparation method of the ternary precursor, a ternary positive electrode material, a secondary battery, a battery module, a battery pack, and an electric apparatus.

Abstract: A modified high-nickel ternary positive electrode material includes an inner core and inner and outer coating layers. The inner core includes a high-nickel ternary positive electrode material matrix, the matrix being doped with M1, M2, and W. M1 is one of Mo, Zr, Ti, Sb, Nb, and Te. M2 is one of Mg, Al, Ca, Zn, and Sr. Chemical formula of the high-nickel ternary positive electrode material matrix is Li1+a[NixCoyMnzM1bM2cWd]O2. 0.65?x<1, 0?y<0.3, 0?z<0.3, 0<a<0.2, 0<b<0.1, 0<c<0.1, 0<d<0.1, x+y+z+b+c+d=1. A surface layer of the inner core is further doped with Co. The inner coating layer is a Co-containing compound. The outer coating layer includes an Al-containing compound and a B-containing compound.

Abstract: Provided are a video recognition method and apparatus, an electronic device, a medium and a computer program product. The video recognition method is described below. A to-be-recognized video is divided into at least two video segments; video frames are extracted from the at least two video segments, and feature recognition is performed on the video frames to obtain initial semantic feature blocks of the at least two video segments; each of the initial semantic feature blocks is fused, and a fused target semantic feature block is obtained; and a type of the to-be-recognized video is determined according to the fused target semantic feature block.

Abstract: This application provides a positive-electrode active material and a manufacturing method thereof. The positive-electrode active material includes a positive-electrode active material matrix and a coating layer, and the coating layer coats a surface of the positive-electrode active material matrix. The positive-electrode active material matrix is Li1+a[NixCoyMnzMb]O2, where 0 < x < 1, 0 ? y < 0.3, 0 ? z < 0.3, 0 < a < 0.2, 0 < b < 0.2, x + y + z + b = 1, and preferably, 0.8 ? x < 1. The element M is selected from more than one of Mg, Ca, Sb, Ce, Ti, Zr, Al, Zn, and B, and the coating layer contains a cobalt-containing compound, an aluminum-containing compound, and a boron-containing compound.

Abstract: A method and an apparatus for training a video recognition model are provided. The method may include: dividing a sample video into a plurality of sample video segments; sampling a part of sample video frames from a sample video segment; inputting the part of sample video frames into a feature extraction network to obtain feature information of the sample video segment; performing convolution fusion on the feature information by using a dynamic segment fusion module to obtain fusion feature information, where a convolution kernel of the dynamic segment fusion module varies with different video inputs; inputting the fusion feature information to a fully connected layer to obtain an estimated category of the sample video; and performing a parameter adjustment based on a difference between the tag of a true category and the estimated category to obtain the video recognition model.

Abstract: The present disclosure provides a method and apparatus for detecting a traffic anomaly, relates to the field of artificial intelligence and specifically to computer vision and deep learning technologies, and can be applied to video analysis scenarios. A specific implementation comprises: acquiring at least two frames of consecutive traffic images; identifying respectively a position of a target vehicle from the at least two frames of consecutive traffic images to obtain a position information set; determining a direction of travel and speed of the target vehicle according to the position information set; and comparing the direction of travel and speed of the target vehicle with a pre-generated vehicle vector field to determine whether the target vehicle is abnormal.

Abstract: Provided are a video recognition method and apparatus, an electronic device, a medium and a computer program product. The video recognition method is described below. A to-be-recognized video is divided into at least two video segments; video frames are extracted from the at least two video segments, and feature recognition is performed on the video frames to obtain initial semantic feature blocks of the at least two video segments; each of the initial semantic feature blocks is fused, and a fused target semantic feature block is obtained; and a type of the to-be-recognized video is determined according to the fused target semantic feature block.