3results about How to "Avoid block effect" patented technology

The invention belongs to the technical field of super-division imaging, and discloses a multi-image super-division imaging method based on the guidance of a frequency domain low-pass filter, and the method comprises the steps: converting a low-resolution source image from a spatial domain to a frequency domain, and obtaining a spectrogram; performing low-pass filtering processing on the spectrogram, and extracting structure and detail information in the source image to generate an initial weight map; positioning a boundary area, and fusing the plurality of source images; estimating the optical flow based on the global motion multi-resolution, capturing direction and displacement information between adjacent pixels in the fused image, and determining the weight of a boundary region according to the direction and displacement information between the adjacent pixels; and finally, performing weighted fusion on the plurality of source images in combination with the initial weight map and the weight of the boundary region to generate a high-resolution image. According to the invention, an efficient super-resolution imaging method is realized, and the generated high-resolution image provides richer and more accurate visual detail information.

This invention discloses an image size adjustment method, apparatus, and storage medium based on model inference. The method includes: generating a summary mapping table; obtaining the data size of an image or video frame; determining whether the image size or video frame size is within a specific range; if within the specific range, determining whether the image size or video frame size is greater than the maximum input size based on the memory and video memory of the execution device and the summary mapping table; if not within the specific range, preprocessing the image or video frame to scale the image size or video frame size proportionally to the specific range, performing data block processing on the compressed image or compressed video frame, performing model inference, obtaining the inference result, and outputting the stitched inference result enlarged to the original image or video frame size before compression. This method can improve super-resolution and restoration effects and has a short processing time.

The invention discloses a multi-focus image fusion and depth recovery method based on a Laplacian pyramid, and the method comprises a Laplacian pyramid construction module which is used for constructing a multilayer Laplacian pyramid expression for an input multi-focus image sequence {Ik} k = 1N (N is the number of images); the multi-focus image fusion module is used for fusing the images at different focus positions based on a fusion strategy of the Laplacian pyramid, and reconstructing a fused image through inverse transformation of the Laplacian pyramid; the focusing degree evaluation module is used for calculating the focusing degree of each image in the image sequence at each pixel position for depth estimation; and the depth recovery module is used for recovering the three-dimensional depth information of the scene based on the focusing degree evaluation result and the servo altimeter data.