335 results about "Pixel matrix" patented technology

The invention relates to the technical field of inspection or monitoring, in particular to an industrial equipment anomaly detection method and system based on machine vision, and the method comprises the following steps: obtaining a multi-channel pixel matrix of an industrial equipment surface image, calculating a pixel gradient and a texture direction change rate, and building a high-dimensional feature space. According to the method, the high-dimensional feature space is constructed by calculating the pixel gradient and the texture direction change rate, so that abnormal feature points are clearly distributed, the misjudgment probability is reduced, the shrinkage factor is dynamically adjusted based on the projection offset, the detection standard is optimized along with data change, and the adaptability is improved; the local gradient change rate and the texture direction change amplitude are combined to carry out offset correction, the precision is enhanced, image block division is subdivided according to structural features, the complex structure recognition capability is improved, local abnormal feature and overall feature relevance calculation is combined with weight correction, the global accuracy is enhanced, sensitive weights are introduced in abnormal point screening, and the judgment precision is improved; and the anomaly detection credibility is enhanced.

The invention discloses an image enhancement method and system based on deep learning, and the method comprises the following steps: obtaining an image pixel matrix, extracting multi-region features according to a brightness difference value, a gradient direction and a texture density partition frequency, carrying out the weighting of the multi-region features, generating a response graph, carrying out the scale decomposition, extracting multi-scale features, fusing a residual image, and enhancing the detail inhibition redundancy. And calling the multi-image parameter to callback the brightness, and outputting an enhanced image. According to the method, accurate separation of high and low frequency regions is realized through region frequency attribute division, fuzzy and excessive enhancement caused by unified processing are avoided, a differentiation strategy is adopted for different regions to improve high-frequency details and suppress low-frequency redundancy, and feature response weight maps are subjected to weighted stacking and normalized enhancement of local textures and contrast. Scale decomposition is combined with multi-scale fusion to improve naturalness and avoid visual discomfort, and edge consistency matching optimizes brightness callback to enhance edge definition and overall visual performance.

The disclosed invention relates to a method for processing image data of an image sensor comprising a matrix of light sensitive elements and a plurality of lens elements and / or filter elements arranged in a pixel matrix in front of a finer sub-pixel matrix of light sensitive elements. A group of light sensitive elements are placed behind a common lens element and / or a common filter element to provide sub-pixel values for a respective position in the pixel matrix. Said image sensor is adapted to capture image data for a plurality of views, wherein each view comprising a matrix of a selected group of sub-pixel values of the pixel matrices captured by the matrix of light sensitive elements.According to the invention, for each captured view of the image, first variations of sub-pixel values in the respective view are determined separately from other views of the same image. Second variations of sub-pixel values related to the same position in the pixel matrix behind a respective lens element and / or filter element are determined in a set of views of the same image. The image data for the image are processed by use of the determined first and second variations.

The invention provides a building main body construction quality tracing method and system, and relates to the technical field of building construction management, and the method comprises the steps: dividing region blocks based on a pixel matrix of a construction site image, carrying out the matching of each region block with a preset control node coordinate in a building information model, recognizing and positioning a target node, generating node image data; based on a target node associated with the node image data, collecting construction environment parameters and physical state parameters of the target node, and generating a multi-dimensional parameter set based on the timestamp index; performing parameter fitting calculation on the multi-dimensional parameter set to generate a parameterized label sequence for representing the construction progress and the quality state; encrypting the parameterized tag sequence and storing the parameterized tag sequence into a distributed account book to form encrypted evidence storage data; and based on the encrypted evidence storage data, quality responsibility tracing and construction record verification are carried out, and a quality tracing verification result is generated. And comprehensive and accurate real-time monitoring on the construction progress and the quality state can be realized.

The invention discloses an image encryption method and system based on a chaotic system, and relates to the technical field of image processing, and the method comprises the steps that the chaotic system is constructed by a cellular automaton and tent mapping combination; performing coordinate scrambling on the coordinate values of the pixel points; all the pixel points form a pixel matrix, and the pixel matrix is traversed to obtain an initial encryption matrix; a pseudo-random sequence is obtained through a chaotic system, the pseudo-random sequence is arranged into a row matrix and a column matrix, the row matrix and the column matrix are subjected to XOR operation, and a plaintext image is converted into a two-dimensional matrix; converting the initial encryption matrix into a two-dimensional encryption matrix; on the basis of a CA model of a cellular automaton, calculating and assigning pixel point values of the initial encryption matrix and the two-dimensional encryption matrix to obtain a target encryption image, and the image encryption method based on the chaotic system can solve the problems that in the prior art, an image encryption method based on the chaotic system is not sensitive to plaintext images, cannot effectively defend differential attacks, is poor in robustness, and cannot effectively defend differential attacks. And the technical problems of periodicity and generally small key space are solved.

The invention discloses a monitoring camera image anomaly detection method and system, and relates to the technical field of intelligent monitoring, and the method comprises the steps: obtaining an original video stream, extracting a pixel matrix of a current frame, carrying out the joint coding of the pixel matrix and a motion vector field of an adjacent frame into a quantum state bit vector, inputting the quantum state bit vector into a pre-constructed anomaly detection operator, and carrying out the detection of the anomaly. Executing quantum state evolution calculation in the Hilbert space, and outputting a two-dimensional distribution matrix of each pixel region; integrating the marked binary image and the physical verification conclusion, and when the quantum anomaly probability value reaches the alarm standard and passes the physical verification, outputting an anomaly alarm signal; and the abnormal alarm signal is converted into a confrontation training sample, a cross-scene virtual sample is generated through quantum noise injection, and an abnormal detection operator is dynamically updated. According to the method, the Hilbert space quantum state evolution step is combined, unitary transformation is applied to the quantum state bit vector, the orthogonal projection operation is executed, the accurately quantified abnormal probability distribution matrix is generated, and accurate assessment of the abnormal risk is achieved.

The invention provides an image encryption method based on chaotic mapping and dynamic positioning generalized knight cruise, and the method comprises the following steps: reading a plaintext image P, and converting the plaintext image P into a pixel matrix with the size of the plaintext image P; a chaotic pseudo-random sequence is generated through three-dimensional Logistic chaotic mapping, and three integer pseudo-random sequences are obtained through integer processing and serve as parameters of a pixel matrix for a knight cruise scrambling method; two of the obtained integer pseudo-random sequences are used for controlling cruising of a rider on rows and columns respectively, so that scrambling operation is completed; and generating a chaos matrix by using a third integer pseudo-random sequence, carrying out XOR operation on the chaos matrix and the scrambled image matrix, changing an image pixel value, and completing image encryption to generate a ciphertext image C. The dynamic positioning generalized knight cruise scrambling method designed by the invention is simple in process, and the encryption efficiency can be remarkably improved.

A method and device for remote sensing image classification using quantum pixel matrix entanglement is provided. The method comprises: preprocessing an acquired raw remote sensing image to obtain image data containing multiband fusion; and calculating a pixel matrix entanglement coefficient μ and a Euclidean distance d between a cluster center and other pixel matrices based on a preprocessed remote sensing image, performing an iterative self-organizing classification on remote sensing image data based on the pixel matrix entanglement coefficient and a Euclidean distance threshold, and obtaining remote sensing image classification results.

The invention discloses a color image encryption method based on deep learning and a four-dimensional Lorentz chaotic system, and belongs to the field of information security and image encryption. According to the method, a CNN deep learning network is constructed, a color image is divided into RGB channels, features of a pixel matrix of each channel are extracted as initial conditions of a chaotic differential equation, and a data set is obtained through solving by a variable-step-size fourth-order Runge-Kutta method and used for training the CNN network. And carrying out first round encryption on the pixel matrix by using a key sequence generated by the trained CNN network, carrying out Arnold scrambling and forward and reverse recursive diffusion on the pixel matrix by using a random chaos sequence generated by an improved four-dimensional Lorentz equation to complete second round encryption, merging the RGB three-channel pixel matrix, and outputting a final color encrypted image. The problems that an existing color image encryption method is low in encryption strength and cannot make full use of color image characteristics are solved.

The invention relates to the technical field of bill text recognition, and discloses a bill text recognition system and method based on deep learning. The method comprises the following steps: acquiring target bill original image data containing a multi-channel pixel matrix and spatial resolution information; based on a matching result of the bill edge features and a preset template, geometric distortion correction is carried out on the original image, and a corrected bill image is generated; inputting the corrected image into a pre-training text region detection network to obtain positioning information containing text line boundary coordinates and region confidence; text line image blocks are extracted according to the positioning information, character segmentation preprocessing is executed, and a character-level image sequence is generated; calling a deep character recognition model to classify the sequence character by character, and generating an initial text recognition result; semantic verification and error correction are performed on the initial result based on the bill type knowledge base, final structured text data are generated, and the processing requirements of bills of different types and qualities can be met.

A method includes receiving, by a computing device, an image indicative of a cross-section of a thermally-sprayed layer. The thermally-sprayed layer defines a porosity comprising a void volume of the thermally-sprayed layer. The image comprises a matrix of pixels, each pixel in the matrix of pixels defining a respective luminance value of a plurality of luminance values. The method includes identifying, based on the luminance values, at least one pixel that is indicative of a void volume in the thermally-sprayed layer. The method includes calculating, based on the at least one pixel that is indicative of the void volume in the thermally-sprayed layer, a total porosity of the thermally-sprayed layer. The method includes determining, by the computing device and based on the at least one pixel that corresponds to a void volume in the thermally-sprayed layer, a quantification of a spatial homogeneity of the porosity of the thermally-sprayed layer.

The present application relates to a kind of for the determination method of liquid alloy density in electrostatic suspension state in space station, after obtaining the photo of liquid alloy in electrostatic suspension state in space station, first photo is converted into pixel matrix and carries out gray scale processing, then deduct photo background by matrix operation.Further, the best binary grayscale threshold of pixel matrix is calculated, and pixel matrix is carried out binary processing.Then all boundary coordinates in pixel matrix are obtained by matrix operation, and each connected boundary is distinguished.Excluding bright spot boundary, the boundary coordinates and center coordinates of liquid alloy are obtained.Then the boundary equation of liquid alloy is obtained by using multi-order Legendre equation fitting liquid alloy boundary coordinates.Finally, the volume of liquid alloy is obtained by rotating integral to the boundary equation of liquid alloy, and then the density value of liquid alloy is obtained.The density value of the present application can be obtained by processing liquid alloy photo, and can be used for the determination of containerless liquid alloy density in electrostatic suspension in space station.

An LCD display unit for a reflection display system in a motor vehicle includes an LCD pixel matrix for selectively controlling picture elements for presenting a display image on a display area; a backlight arrangement, which is divided into selectively controllable dimming zones in order to illuminate regions containing multiple picture elements with background light; and a control unit, which is configured to control the LCD pixel matrix so that the display image is output, the display image being distorted using a warping function in order to compensate for a distortion of the display image as a result of reflection from a reflection region. The dimming zones have a shape and a size determined by the warping function.

The invention discloses a real-time image super-resolution processing system and method based on an FPGA, and relates to the technical field of image processing, and the system comprises at least one convolution module and a pixel alignment and rearrangement module. The convolution module obtains convolution input pixel data, edge pixel filling processing and caching processing are carried out on the convolution input pixel data to obtain a pixel matrix, convolution processing is carried out according to the pixel matrix to obtain convolution output pixel data, and data caching is carried out based on the block random access memory in the processing process of the convolution module; the pixel alignment and rearrangement module obtains to-be-aligned pixel data, alignment processing and rearrangement processing are carried out on the to-be-aligned pixel data, processed target output pixel data are obtained, and the to-be-aligned pixel data is convolution output pixel data output by the last convolution module; the resolution of the target output pixel data is higher than that of the to-be-processed image frame. In this way, the efficiency of image super-resolution processing is improved, and the real-time requirement is met.

The invention relates to the technical field of video processing, in particular to a target area video compression quality evaluation method based on video dynamic semantic segmentation. The method comprises the following steps: acquiring an original video sequence; performing pixel-level semantic segmentation on each frame of the original video sequence by using a pre-trained video semantic segmentation model, and outputting a binary target area mask; performing compressed encoding on the original video sequence, and outputting a compressed video sequence; according to the binary target area mask, extracting a pixel matrix of a target area in the original video frame and the compressed video frame; and calculating the PSNR of the original video frame and the compressed video frame in the target area pixel by pixel based on the pixel matrix, and solving the average value of the PSNR values of all pixels to obtain the average PSNR value of the single-frame target area. According to the method, quality degradation during scene switching or strenuous movement is accurately captured, and the evaluation sensitivity of the target area is improved compared with that of a traditional method.

The invention relates to the technical field of intelligent peripherals, in particular to a user-oriented interaction information expansion method, which comprises the following steps that: each keycap acquires displacement and pressure, generates an event packet, uploads the event packet through an optical wireless link, and obtains a clock and supplies power; constructing a keycap graph by a master controller, obtaining an intention vector through graph convolution and pooling, inputting the intention vector into a cyclic network to generate a pixel matrix and a phase matrix, and downlink after Bernoulli sparse coding; performing key cap side orthogonal matching pursuit decoding and four-round gradient optimization, and outputting a differential pixel set, a differential phase set and a synchronization mark; the master control refreshes the digital micromirror array and the ultrasonic phased array in a differential manner, locks a frame-level time sequence through a synchronization pulse, compensates link delay and hardware drift in real time based on Kalman filtering, and keeps an optical-acoustic alignment error to be less than 100 microseconds; degradation monitoring enables a link to be automatically switched to a pre-carved character layer and to be quickly recovered when the link is abnormal, and a keyboard interaction information expansion scheme with low delay and high consistency is provided.

The invention provides a musculoskeletal X-ray image analysis method, which comprises the steps of converting an initial judgment result of a musculoskeletal X-ray image to be analyzed into a one-dimensional tensor, and obtaining a group of prediction probability values Prob1 through probabilistic processing; converting the pixel matrix of the image data of the musculoskeletal X-ray image to be analyzed into a three-dimensional tensor representing the number, height and width of image channels; constructing a DenseNet121 model, and inputting the three-dimensional tensor into the trained DenseNet121 model to obtain a group of first image feature representations; constructing a ResNet50 model, and inputting the three-dimensional tensor into the trained ResNet50 model to obtain a group of second image feature representations; and splicing the prediction probability value Prob1, the first image feature representation and the second image feature representation to obtain feature data for representing the musculoskeletal state. According to the method, a mixed judgment model in which human participates in deep learning is designed, and doctors can be accurately assisted in judging musculoskeletal states.

The invention discloses an intelligent glass fiber reinforced plastic curing temperature control method and system in combination with visual detection, and relates to the field related to intelligent manufacturing, and the method comprises the steps: carrying out the collection of a morphological spectral image and an infrared thermogram of a glass fiber reinforced plastic curing surface through a multispectral visual detection sensor; establishing a multi-dimensional pixel matrix including surface texture, reflection spectrum and infrared temperature distribution; carrying out polymerization analysis on optical coherence and temperature distribution continuity, constructing a photo-thermal coupling consistency tensor, and adaptively generating a curing region partition map; calculating a temperature gradient vector field of each adjacent region, and constructing a control mapping function by taking a morphological response characteristic index as a weight parameter; and a differential heating instruction is configured to control the multi-point array heating unit to execute self-adaptive heating control. The technical problem that existing glass fiber reinforced plastic curing temperature control cannot comprehensively and accurately sense and dynamically adapt to complex changes of the curing process is solved, and the technical effect of intelligent, accurate and differential control over the glass fiber reinforced plastic curing temperature is achieved.

The embodiment of the invention provides a light spot feature calculation method, device and equipment for a dynamic image, and is applied to the technical field of optical image processing, and the method comprises the steps: carrying out the summation of each row of a non-zero pixel matrix, obtaining a column vector, multiplying the column vector by an index row vector, dividing the product sum by the sum of non-zero pixels, and obtaining the light spot feature of the non-zero pixel matrix. Obtaining a y coordinate value; summing each column of the non-zero pixel matrix to obtain a row vector, multiplying the row vector by the index column vector, and dividing the product sum by the sum of the non-zero pixels to obtain an x coordinate value; according to the x coordinate value and the y coordinate value, x-direction centroid coordinates and y-direction centroid coordinates of the light spot are obtained; calculating the variance of the x direction and the y direction according to the x coordinate value and the y coordinate value, and calculating the light spot roundness and the light spot diameter based on a least square method according to the variance of the x direction and the y direction; and calculating a beam waist radius and a beam divergence angle based on the light spot diameter. In this way, the real features of the light spots in the dynamic image can be accurately calculated.

The invention discloses a quick color image encryption method based on a four-dimensional discrete hyper-chaotic system. The method comprises the following steps of: disturbing initial condition parameters through a hash value obtained by a plaintext image and an external random character string to obtain disturbed initial condition parameters; substituting the four-dimensional discrete hyper-chaotic system for iteration to obtain four chaotic sequences; carrying out compressed sensing and two-dimensional discrete wavelet packet transformation on the plaintext image to obtain a to-be-encrypted pixel matrix; rearranging the chaotic sequence; performing index scrambling on the matrix m1 and the matrix m3 by adopting the matrix m2 and the matrix m4 to obtain a matrix M1 and a matrix M2; scrambling the M1 by adopting the M2 to obtain a matrix M3; scrambling a to-be-encrypted pixel matrix by adopting the matrix M3 to obtain a pixel matrix P4; and exporting a pulse sequence and coordinates for diffusion from the chaos sequences s2, s3 and s4, and sequentially acting on a pixel matrix P4, thereby obtaining a final ciphertext image. According to the method, the color image can be effectively encrypted, and the method has good security for resisting various attacks.

The embodiment of the invention discloses an object tracking method, device and equipment and a computer readable storage medium, and the method comprises the steps: determining a target frame image after a current frame image from a target video sequence, and enabling the current frame image to comprise a plurality of to-be-tracked objects; based on a target detection model and a target segmentation model, determining a first mask corresponding to the target frame image, the target detection model being a model having zero sample recognition capability and language recognition capability, and the mask being a pixel matrix corresponding to the target frame image; determining a second mask corresponding to the target frame image based on the target tracking model and the target frame image; determining a target mask corresponding to the target frame image based on the first mask and the second mask; and based on the target tracking model, the target mask and the target frame image, obtaining each to-be-tracked object in the target frame image and a tracking result of the target object, and the target object being a newly added object except the plurality of to-be-tracked objects in the target frame image.

The invention discloses a pixel extraction and conversion method of an LEN format image, and belongs to the technical field of digital image processing. In order to solve the technical problems that the LEN format has no public specification and general software cannot read data, binary analysis is carried out through an LEN sample, and the core binary layout of a file is found for the first time: the image width and height are stored in specified offset positions, and a small-end byte order is adopted; pixel data comprises a compressed data segment and a row information segment, and an RLE coding rule is adopted. The method comprises the following steps: acquiring an LEN format file; analyzing the file header to obtain an image size; performing segmented analysis on the image data; synthesizing the analyzed data into image data; and packaging the pixel matrix into a standard TIFF format file. The LEN core binary layout is disclosed for the first time, zero distortion and cross-platform conversion are achieved, reading and storage of screening parameters are not involved at all, existing patents are effectively avoided, and the method can be used for preview, filing and third-party system integration in the flexographic printing process.

A method includes receiving, by a computing device, a raw image indicative of a cross-section of a thermally-sprayed layer. The image includes a matrix of pixels, and each respective pixel in the matrix of pixels defines a luminance value. The method may further include determining, based on the luminance values, at least one pixel that corresponds to an oxide component in the layer and removing the at least one pixel that corresponds to the oxide component in the layer to generate a modified matrix of pixels. The method may further include generating an oxide-filtered image based on the modified matrix of pixels. The method may further include converting, by the computing device and based on the luminance values, the oxide-filtered image into a binary image and determining, by the computing device and based at least partially on the binary image, a porosity of the coating layer.

The application discloses a high-uniform seamless splicing system and a brightness adjusting method for light field display, and belongs to the technical field of light field display. The system uses a minimum display unit modular splicing architecture to realize super-high resolution display, adopts a composite microlens array to realize axial displacement of an imaging surface, covers a gap between adjacent display units through optical expansion of a pixel matrix, and eliminates parallax faults caused by physical splicing seams. The system integrates a distributed dynamic uniform light module, collects brightness data of a display area in real time, performs brightness compensation on a low-illumination area through array light source driving control, changes a light transmittance, and regulates and controls pixel-level brightness uniformity of the system. The scheme effectively solves the problems of pixel continuity loss and adjacent dimming area light coupling in high-resolution light field reconstruction. The system is widely applicable to light field three-dimensional display, and is a scheme for improving three-dimensional display effect.

The invention provides a wheat disease identification method and device, and is applied to the technical field of crop disease identification. The method comprises the steps of obtaining a training sample set; determining mass-to-charge ratio signal intensity data of volatile organic compounds of the wheat leaf samples in the training sample set, determining a maximum characteristic value of each mass-to-charge ratio channel in the mass-to-charge ratio signal intensity data, mapping the maximum characteristic values of all the mass-to-charge ratio channels into pixel points, and obtaining a two-dimensional pixel matrix of the wheat leaf samples; extracting parent ions and pixel points where fragment ions corresponding to the parent ions are located from the two-dimensional pixel matrix, and rearranging the positions of the extracted pixel points in the two-dimensional pixel matrix to obtain a recombined feature map; and inputting the recombinant feature map into a wheat disease recognition model for network training, wherein the wheat disease recognition model is used for recognizing the disease type and infection degree of wheat according to the mass-to-charge ratio signal features in the recombinant feature map.

A LIDAR device for scanning a field of vision with the aid of an emitter module that includes an emitter pixel matrix made up of emitter pixel columns and emitter pixel rows and with the aid of a detector module that includes a detector pixel matrix made up of detector pixel columns and detector pixel rows. Individual or all emitter pixels of the emitter pixel matrix and / or individual or all detector pixels of the detector pixel matrix is / are selectively activatable.

A CMOS image sensor includes a unit pixel array including a photodiode array, a color filter array, a micro-lens array, and a grid isolation structure laterally separating adjacent color filters. The grid isolation structure includes a first low-n grid, a second low-n grid underlying the first low-n grid, and a metal grid within the second low-n grid, the first low-n grid being narrower than the second low-n grid. The color filter array includes color filter matrixes, all color filter matrixes have the same arrangement pattern. Sizes of color filters in each color filter matrix vary depending on locations of the color filters in the color filter matrix. In an edge portion, a distance between a center of a color filter matrix and a center of a corresponding unit pixel matrix in plan view varies depending on a location of the unit pixel matrix in the CMOS image sensor.

The invention discloses a method for calculating large-area displayable graphic and text information under a point light source. The method comprises the following steps: S1, digitizing the graphic and text information; s2, point light source parameter modeling; s3, carrying out basic modeling on the microscopic fringes; s4, stripe width correction is carried out in combination with an image-text information model; and S5, carrying out block calculation and splicing on the large-area region. By digitally processing the graphic and text information, the complex graphic and text can be converted into an accurate pixel matrix, a clear data basis is provided for subsequent calculation, and the accuracy of the graphic and text information at a processing source is ensured; point light source parameter modeling is carried out, the position, illumination intensity and other key parameters of a point light source can be accurately captured, a reliable optical basis is provided for subsequent microcosmic fringe design, and calculation deviation caused by fuzzy light source parameters is avoided. And a brand new microscopic stripe basic model is constructed, so that stripe distribution can more flexibly adapt to point light source characteristics and graphic and text information requirements.

The invention relates to the field of image processing, in particular to a laser marking positioning method and system based on vision, and the method comprises the steps: obtaining a template RGB image and a to-be-positioned RGB image of a bottom plate of an electric iron; constructing a pixel matrix set of the target hole, and calculating the distinction degree between the target hole and any other hole; calculating the positioning confidence of the target hole under the target scale, traversing to obtain the positioning confidence of the target hole under each scale, and determining a positioning hole according to the positioning confidence; calculating the positioning weight of any pixel matrix in the pixel matrix set of any positioning hole, and taking any steam hole in the RGB image to be positioned as a hole to be matched so as to calculate the positioning matching degree of the hole to be matched and any positioning hole at the target scale, and calculating the comprehensive matching degree of the to-be-matched hole and any positioning hole based on the positioning matching degree and the positioning confidence accumulated value. According to the technical scheme, the marking and positioning precision can be improved.

The application relates to the fields of information security and image processing, and discloses a multi-image parallel encryption processing method based on an improved chaotic mapping, which comprises the following steps: separating and stacking each color channel of multiple images to be encrypted to construct a three-dimensional pixel matrix; extracting a matrix hash digest to be converted into a sub-key; inputting the sub-key into an improved chaotic mapping model of a historical state variable to iteratively generate a scrambling and diffusion sequence; performing sorting scrambling on the three-dimensional matrix in multiple orthogonal dimension sections by using the sequence; mapping the sequence into a shift step length to perform a cyclic shift and a bitwise XOR linkage diffusion on the pixels; and finally, outputting ciphertext images by segmentation. The application breaks the physical isolation of a single image to realize spatial depth mixing by constructing a three-dimensional matrix and performing cross-dimension scrambling; the chaotic dynamics degradation under limited precision is overcome by means of a historical state delay; and the efficiency of multi-image encryption and the system attack resistance are improved by using a shift and XOR linkage to realize bit-level depth confusion.