31results about How to "Enhance edge information" patented technology

The invention discloses an HDR image fusion method based on a plurality of LDR images with different exposure degrees. The method comprises the following steps: step one, carrying out dark-bright region division on a plurality of inputted LDR images with different exposure degrees and calculating weight maps of the images according to contrast ratios, saturability values, exposure degrees, and brightness of the images in all regions; and step two, carrying out image decomposition by using a gradient pyramid and carry out image fusion in a multi-resolution manner. Compared with the prior art, the method has the following beneficial effects: because of usage of the gradient pyramid instead of a Laplacian pyramid during the fusion process, the image direction and edge information is increased and thus the generated HDR image effect is good; the image brightness information is considered during weight solving, so that the generated HDR image has the bright color, high contrast ratio, and clear image details; and the HDR image fusion method has the good referential significance in the HDR image field.

The invention discloses an intelligent monitoring system for a power cable production process. The system comprises an image acquisition assembly, a sensor assembly, a wireless transmission module, anintelligent monitoring center and a danger alarm module. The sensor assembly is used for collecting environmental data and temperature data of the surface of power cable production equipment And transmitting the acquired data to the intelligent monitoring center; the intelligent monitoring center processes the received data and compares the processed data with a preset safety threshold value; andwhen the data is higher than a preset safety threshold value, the danger alarm module gives an alarm, image acquisition is carried out on the position where the sensor node for acquiring the data islocated through the image acquisition assembly, and the acquired image is displayed after being processed. The beneficial effects of the invention are that the system achieves the effective monitoringof the environment of a power cable production workshop and the operation state of power cable production equipment, and improves the monitoring intelligence of a power cable production process.

The invention discloses a touch screen-oriented finger outer skeleton wearable force-touch sense interaction device, which can provide multiple modes of force sense and vibration touch sense feedbacks for human fingers. A plane four connection rod mechanism and a direct-current motor are used as an execution mechanism to provide continuous controllable force sense feedback for an operator. Different types of touch sense feedbacks are provided for the operator by controlling vibration amplitude and frequency of a voice coil vibrator, and the authenticity of touch sense perception can be enhanced in an interaction process; a vibration motor can enhance edge information of an image, and can achieve a collision detection effect. Finally, three-dimensional feature information of a virtual object in a touch screen, such as texture, an outline shape, and roughness can be expressed. The interaction device has important meaning in improving the authenticity and the fidelity of force touch sense feedbacks.

A pancreatic cystic tumor image classification method based on multi-channel multiple classifiers comprises the following steps: 1) performing window width and window level adjustment on an original image, and performing Canny edge detection and gradient amplitude calculation to enhance edge features; 2) adopting a ResNet to perform end-to-end training on a multi-channel graph, using an output ofa pool5 layer as extracted features, using a Bayesian classifier and a KNN classifier to perform classification, and obtaining the classification probabilities; and 3) using a random forest classifierto classify the obtained 3 different probabilities to get a final result, wherein a random forest is composed of multiple decision trees. The pancreatic cystic tumor image classification method basedon multi-channel multiple classifiers can automatically perform edge enhancement, and can improve classification accuracy.

The invention discloses a time domain high-pass filtering method. The method includes the steps of carrying out preliminary filtering on an original infrared image by using a mask plate, obtaining a difference value of the two adjacent frames of infrared images after preliminary filtering to obtain a difference value image, carrying out bilateral filtering on the infrared image after preliminary filtering, obtaining a residual image after bilateral filtering, enabling the part with the gray value greater than a gray threshold value in the difference image to belong to a fast moving area, enabling the part, with the gray value smaller than or equal to the gray threshold value, of the difference image to belong to a slow moving area, determining correction items of the fast moving area and the slow moving area according to the residual image after bilateral filtering and different time constants, and correcting the original infrared image through the correction items to obtain a corrected infrared image. According to the invention, the non-uniformity of the infrared image is effectively reduced, the occurrence of ghost images is inhibited, and the correction effect is good.

The invention belongs to the technical filed of computer vision and especially relates to brightness enhancement and edge enhancement of a grayscale image. The method specifically comprises the following steps: estimating global brightness and determining algorithm self-adaption parameters, generating a brightness mapping graph of the images and carrying out calculation to obtain a brightness enhancement image and carrying out edge enhancement processing. The method is characterized by, to begin with, estimating the self-adaption parameters according to the brightness distribution conditions of global dark areas; then, carrying out global brightness enhancement processing on the images and obtaining the modulation mapping graph of the whole picture through a modulation function, and carrying out calculation to obtain brightness enhancement result; and finally, realizing edge enhancement based on a dimension-self-adaptive Gaussian difference model, the model dimension being influenced by contrast ratio, and therefore, bright areas can be enhanced with finer texture information, and the dark areas can be enhanced with larger contour information. The method can enhance overall brightness and contrast of the grayscale images effectively, and the self-adaption characteristic can play a very good effect on edge enhancement of the bright and dark areas.

The invention discloses an automobile glass sub-pixel contour extraction method and an automobile glass detection method and device, belongs to the technical field of glass detection, and is used forsolving the problems of large measurement error, long detection time and the like of an existing contact type glass detection method. The adopted scheme is specifically as follows: acquiring a standard automobile glass image and an automobile glass image to be detected; preprocessing each automobile glass image, wherein preprocessing comprises Sigma filtering, median filtering and image enhancement; carrying out edge extraction on each preprocessed automobile glass image to obtain a pixel-level edge contour of each automobile glass image; performing sub-pixel positioning on the pixel-level edge contour to obtain a sub-pixel edge contour of each automobile glass image; registering the obtained standard automobile glass outline and the outline of the automobile glass to be detected; and calculating the error size of the to-be-detected glass. The method and the device have the advantages of high registration precision, high detection efficiency, high detection precision and the like.

The present disclosure provides geological linear body extraction method based on tensor voting coupled with Hough transformation, including pre-processing a remote sensing image to obtain a pre-processed remote sensing image; selecting three optimal wavebands from N multi-spectral wavebands of the pre-processed remote sensing image, so as to obtain a remote sensing image combined by the optimal wavebands, N being a natural number greater than or equal to 3; using Gaussian high-pass filtering to perform sharpening processing on the remote sensing image combined by the optimal wavebands, so as to enhance linearized edge information; performing edge detection on the remote sensing image having enhanced linearized edge information, so as to obtain all edge points in the remote sensing image; and converting all the edge points in the remote sensing image from an image coordinate system to a parameter coordinate system, and extracting a geological linear body from the parameter coordinate system.

The invention discloses a method for enhancing the edges of synthetic aperture radar images. The method includes computing each pixel in an original synthetic aperture radar image to obtain normalized average ratios of the pixels; using the normalized average ratio of each pixel in the original synthetic aperture radar image as a gray value of the pixel and redrawing a synthetic aperture radar image. The method has the advantages that the average ratios of a sequence are optimized to obtain the gray values of corresponding elements, so that the synthetic aperture radar image can be redrawn, and the strength of the edges of the synthetic aperture radar image is enhanced.

The invention provides a method for automatically generating a tetragonal continuous pattern based on content features. The method is characterized in that firstly, the noise of a fractal image is filtered by a smoothing filter algorithm, and grayscale is carried out; then the edge distribution histogram of gray-scale image is detected and extracted by Canny, and the probability distribution of four edge elements is calculated to judge whether the fractal image needs to be extracted or not, then the content of the fractal image is retrieved and the pattern position is determined by using the rectangular window rotation mechanism, and finally, the quadrangle continuous pattern is generated by calling horizontal and misalignment methods, so that the pattern generation method of the inventionautomatically extracts patterns and generates tetragonal continuous patterns through the main steps of filtering, edge feature retrieval, pattern content positioning, pattern extraction and stitching, and the like.

The invention discloses a target detection method based on an SSA sharpening attention mechanism, in the method, user-defined sharpening filtering is introdued into a space attention module for the first time, and combined with a channel attention module for use, thereby reducing the influence of interference factors on the SSA sharpening effect, and serving the sharpening effect; according to the invention, the edge information of the detected object in the neural network is enhanced spatially, and object positioning is enhanced. Edge information of large objects can be perfected, existence of small and medium objects in an output layer can be improved, and the detection effect is improved; according to the invention, the combination mode and the embedding position of the SSA space sharpening module and the channel attention module are perfected. Compared with a space attention module in the CBAM, the effect on the lightweight target detection model is better. According to the SSA space sharpening module, the required calculation amount and parameter amount are extremely small, the detection speed is hardly influenced, and the lightweight module is high in practicability, plug-and-play and easy to implement.

The invention discloses a calculation method for SAR image coherent speckle suppression with optimal diffusion coefficient, comprising the following steps: calculating εR(x, y) of the point matrix of the SAR image; calculating the diffusion coefficient c(g); Perform speckle suppression. The improved diffusion coefficient of the present invention has improved the inconsistency of diffusion speed when the existing diffusion coefficient is in the homogeneous region; since the diffusion coefficient of the present invention can realize isotropic diffusion in the homogeneous region, i.e. cη=cξ=1, it can be better In the heterogeneous area, when cη=cξ=0, the diffusion is stopped, which can inversely smooth the edge of the SAR image and enhance the edge information of the image. The ratio of the diffusion rate in the η direction to the diffusion rate in the ξ direction is zero, and it can be seen that the diffusion coefficient satisfies the Charbonnier criterion. Experimental comparison verifies the validity and superiority of the calculation method of the present invention, and theoretically proves that the diffusion coefficient of the present invention is iteratively stable in the coherent speckle suppression process.

The invention discloses a method for measuring base curvature data of goggles based on medical images. , according to the image data, construct the 3D head skin model of the medical image; extract the 3D contour of the 3D head skin model according to the volume data field; convert the 3D contour into a symmetrical 2D contour, and measure the base Bending parameters; calculate the base bending value based on the base bending parameters. By adopting the embodiments of the present invention, the measurement accuracy and efficiency of base curvature data of asymmetric medical images can be improved.

The invention provides a composite material terahertz imaging resolution enhancement method and device and equipment and a medium. The method comprises the steps: carrying out the imaging detection of a detected composite material through a terahertz imaging system, and obtaining a first one-dimensional range profile of the detected composite material; denoising the first one-dimensional distance image to obtain a denoised second one-dimensional distance image; respectively taking the second one-dimensional distance image and the detail-enhanced second one-dimensional distance image as a first guide image and a second guide image, and respectively carrying out guide filtering on the second one-dimensional distance image to respectively obtain a detail layer and a base layer; performing first gain and second gain on the details and the base layer respectively, and superposing the detail layer subjected to the first gain and the base layer subjected to the second gain into a third one-dimensional distance image; and performing amplitude imaging on the third one-dimensional range profile to obtain an imaged result image. According to the method provided by the embodiment of the invention, the detail information such as the edge of the image can be enhanced while the background noise is effectively suppressed.

The invention relates to the technical field of image recognition, in particular to a distance-adaptive thermal infrared face recognition method, which comprises the following steps: constructing and training an improved thermal infrared image super-resolution enhancement network Retinex-CNN, and processing a thermal infrared image by using the trained improved thermal infrared image super-resolution enhancement network Retinex-CNN; on the basis of prior information obtained by near-infrared, different feature algorithms are utilized, local feature extraction is carried out on near-infrared and processed thermal infrared images, full feature fusion is carried out on the extracted different features, dimension reduction processing is carried out on the fused features, and then classification and recognition are carried out on the features after dimension reduction. Through the recognition method, the problem that face recognition is difficult due to the fact that thermal infrared imaging is greatly influenced by the distance when the distance changes can be effectively solved.

The invention discloses a feature extraction method for a low-quality finger vein image, and the invention introduces a displacement replacement method for the displacement of an image in consideration of the severe deviation of the finger vein image and the loss of part of information caused by the direct cutting of the image in a preprocessing link. Considering the defects of a local mode methodand a vein network method, the invention provides superposition extraction of PBBM features on the basis of MCM feature extraction of an image. Therefore, clearer finger vein features can be extracted under the condition that full-image information is reserved, and the recognition accuracy of a low-quality image is effectively improved.