9 results about "3D lookup table" patented technology

This invention provides a method for extracting three-dimensional features of steel stamps on water transfer printing paper based on multispectral fusion, relating to the fields of optical inspection, machine vision, and automated quality control. This method utilizes adaptive illumination and image acquisition, employing optimal band optimization, polarization decoupling, and dynamic structured light projection to obtain a series of images separating geometric deformation and pattern interference. Three-dimensional reconstruction is performed using a physically constrained neural network, embedding a double-layer reflection optical model into the network to decouple and predict the paper substrate normal map and integrate to generate a high-precision three-dimensional point cloud. Multi-dimensional geometric features such as height, curvature, and shape are then extracted from the point cloud. Finally, based on a three-dimensional lookup table, the features are mapped to a comprehensive defect level, and automated, differentiated graded processing is performed. This invention achieves non-contact, micron-level high-precision inspection and intelligent quality control of steel stamps on complex surfaces.

The application discloses a three-dimensional look-up table (3D-LUT) construction method and a color correction method based on FPGA, wherein the construction method comprises: interval sampling a 3D-LUT from three dimensions to form a three-dimensional look-up table, and the three-dimensional look-up table comprises: eight Sub-LUTs for storing split 3D-LUT data; wherein N is the sampling point number of three RGB channels in the 3D-LUT; the 3D-LUT contains (N-1) 3 cubic units; the Sub-LUT contains (M-1) 3 cubic units; and M is the sampling point number of three RGB channels in the Sub-LUT. The method can reduce BRAM resource occupation and is suitable for the technical field of image and video processing.

The application discloses an image processing method and device and a training method, relates to the field of image processing, and particularly relates to the fields of computer vision, deep learning and artificial intelligence. The specific implementation scheme is as follows: for each pixel point in a to-be-processed image, according to a first color value of the pixel point, a voxel index of the pixel point in a three-dimensional lookup table and a relative coordinate of the pixel point in a target voxel corresponding to the voxel index are determined; according to the voxel index, first color information of each vertex of the target voxel is acquired; a weight generation model obtained by joint training of the three-dimensional lookup table is used to determine a first weight of the vertex according to the relative coordinate; according to the first color information and the first weight, second color information of the pixel point after color mapping is determined; and a target image is generated according to the second color information of each pixel point.

The present disclosure provides a method for generating a screen color calibration file and a screen color calibration method, and relates to the field of screen color calibration under multiple luminances. The method comprises the following steps: determining a color coordinate matrix of three primary colors corresponding to the screen under N luminances, wherein N is a positive integer greater than 2; determining a conversion function matrix corresponding to the screen according to the N color coordinate matrices, wherein the conversion function matrix is used to map the color gamut of the screen under each luminance to the color gamut under a reference luminance; performing color calibration on the screen under the reference luminance to obtain a three-dimensional lookup table under the reference luminance; and determining the conversion function matrix and the three-dimensional lookup table as the color calibration file of the screen. Thus, the color gamut under different luminances can be kept consistent, and color calibration under the maximum luminance can ensure the color accuracy under all luminances, thereby ensuring the reliability of screen color calibration and improving the calibration efficiency.

This invention discloses a real-time downconversion method for high dynamic range to standard dynamic range, comprising the following steps: S10, inputting an HDR video image into a static 3D lookup table (3DLUT) downconversion algorithm template, and outputting a first standard dynamic range (SDR) signal; S20, inputting the first SDR signal into an SDR image detection module to determine whether there is overexposure abnormality in the image after 3DLUT downconversion: if there is no overexposure abnormality, directly outputting the first SDR signal; if there is overexposure abnormality, correcting the 3DLUT mapping curve through a 3DLUT mapping curve adjustment module to complete the static 3DLUT parameter update; then, based on the updated static 3DLUT, regenerating the SDR signal through the static 3DLUT downconversion algorithm template, repeating the above detection and static 3DLUT parameter update process until the output SDR signal image has no overexposure abnormality.

The application relates to a backlight white balance control method and device, electronic equipment and a storage medium. The method comprises the following steps: constructing a multi-temperature three-dimensional lookup table set; detecting the current working temperature of a backlight module when an image is displayed on a backlight display device, and determining the target light output characteristics of the current display picture; selecting a target three-dimensional lookup table from the multi-temperature three-dimensional lookup table set according to the current working temperature; inputting the target luminance value and the target chroma coordinate into the target three-dimensional lookup table to obtain a target three-color driving current value and generate a driving signal, and driving the backlight module to output mixed light matched with the target light output characteristics, so as to realize backlight white balance control. Based on the luminance, chroma and three-color driving current mapping relationship of the backlight, the mixed light meeting the target white balance requirement can be quickly and accurately output, the chroma coupling error caused by the traditional independent correction of the sub-channel is avoided, and the control precision and response speed of the backlight white balance are improved.

A device includes a memory circuit capable of storing a non-uniform 3-dimensional lookup table (3D LUT) specifying color correction data. For one or more of a red dimension, a green dimension, and a blue dimension, the non-uniform 3D LUT includes a number of color correction values within one or more first light ranges of the non-uniform 3D LUT that exceeds a number of color correction values included within one or more second light ranges of the non-uniform 3D LUT. The device includes a color correcting circuit capable of transforming a source frame into a color corrected frame using the non-uniform 3D LUT.

This invention discloses a multi-layer cloud detection method based on a 3D lookup table. The method constructs and trains a 3D lookup table in the model training phase, and constructs a confidence score based on the confidence parameters retrieved from the trained 3D lookup table in the model invocation phase. The final confidence score determines the probability of a multi-layer cloud. The model training phase includes a data feature creation stage, a feature combination stage, a 3D lookup table creation stage, and an optimal segmentation threshold determination stage. The model invocation phase includes a 3D lookup table reading stage, a threshold segmentation stage, and a weight combination stage. This invention solves the problems of low accuracy and poor linearity between the final confidence score and detection accuracy in existing technologies. The method of this invention can be used 24 / 7, both day and night. Detection results show good linearity between the final confidence score and detection accuracy, with high accuracy at a final confidence score of 1, and a maximum detection accuracy of approximately 90%.