217 results about "Adaptive histogram equalization" patented technology

The invention discloses an unmanned aerial vehicle electric power inspection image intelligent analysis method and system based on deep learning and multi-modal fusion and a medium thereof, and relates to the technical field of electric power equipment detection. The method comprises the following steps: planning an optimal inspection path by adopting an A * algorithm to realize multi-sensor synchronous data acquisition; adaptive histogram equalization and defogging processing are carried out on the visible light image, non-uniformity correction and temperature calibration are carried out on the infrared image, and filtering and registration are carried out on point cloud data; constructing a multi-scale feature fusion network based on improved VGGNet-16, and introducing deformable convolution and a cross-modal attention mechanism to realize multi-source data fusion; defect detection is carried out based on a three-level template library and a feature map cross-correlation algorithm, and the precision is improved in combination with non-maximum suppression and sub-pixel positioning; and finally generating a detection report containing defect types, positions and maintenance suggestions. According to the invention, the automation level and the detection precision of power inspection are obviously improved.

The invention discloses a method for identifying impurities and bubbles in a wine bottle based on multi-feature comprehensive judgment, and belongs to the technical field of image processing and computer vision, and the method comprises the steps: employing a high-resolution industrial camera to be combined with an LED light source, and continuously shooting multiple frames of images of the wine bottle under the consistent shooting angle and illumination condition; carrying out the preprocessing of the collected image through employing a self-adaptive histogram equalization and Gaussian blur method; extracting the edges of impurities and bubbles in the wine bottle on the image by adopting a Canny edge detection algorithm to obtain an edge image; segmenting the edge image into a plurality of target areas by using a contour detection technology, and extracting possible bubble and impurity areas from the edge image; extracting target area features including circularity, gray average, gray uniformity, texture features and color space; and according to the extracted features, bubbles and impurities are distinguished by applying a comprehensive judgment rule. According to the method, the bubbles and the impurities are comprehensively judged by adopting multi-feature combination, so that the accuracy and the robustness of detection are greatly improved.

The invention discloses a cancer image denoising method and system based on a multi-scale neural network, and the method comprises the steps: carrying out the adaptive segmentation of a cancer image, carrying out the noise estimation and statistical analysis of each segmented region, and automatically adjusting the denoising parameters; performing multi-scale feature extraction by adopting wavelet transform and pyramid decomposition, and analyzing and processing the image after parameter adjustment; performing quality evaluation and distortion detection on the image after multi-scale processing, automatically identifying and positioning artifacts and distortion, and correcting and compensating interference by adopting a point spread function inversion and Wiener filtering method to obtain a corrected image; and after histogram equalization and contrast-limited adaptive histogram equalization are applied to the corrected image, denoising processing is carried out by using a convolutional neural network in combination with sparse representation and a dictionary learning algorithm, and a denoising result is obtained. According to the method, noise in different types of cancer images can be effectively removed, meanwhile, key diagnosis information is reserved, and the image quality and definition are improved.

The invention discloses an industrial machine vision recognition and analysis system, relates to the technical field of machine vision recognition, and is used to solve the problem of poor industrial machine vision recognition. The invention combines total variation denoising, contrast-limited adaptive histogram equalization, random geometric transformation and other technologies to enhance image diversity, improve the system's generalization ability for images in different environments, extract image texture and edge features, provide high-quality data for subsequent analysis, use unsupervised learning to extract image potential features and patterns, and locate abnormal areas through template matching, so as to adaptively discover defects in the production process, dynamically update the model based on a small amount of labeled data, optimize feature representation in real time, and adapt to environmental changes. Finally, semi-supervised learning and pseudo-label generation technology are used to expand the training set, improve the generalization ability of the model, maintain efficient learning when labeled data is limited, reduce manual labeling costs, and improve the accuracy of automated quality inspection.

The invention relates to a hydraulic engineering concrete crack intelligent identification and quantitative analysis method and system based on machine vision. The method comprises the following steps: firstly, acquiring an original image sequence of the surface of a hydraulic engineering concrete structure, classifying according to illumination intensity, shooting angle and shooting distance, extracting crack edge features through a convolutional neural network, and fusing to obtain a crack feature set; correcting illumination through adaptive histogram equalization, correcting angles and distances through geometric transformation, and combining edge detection and scale invariant feature transformation to obtain standardized geometric parameters including crack length, maximum width and the like; if the parameter exceeds the engineering safety standard threshold value, tracking a crack track through an optical flow method to calculate increment, and inputting a neural network to output a damage trend; and finally, generating a three-color risk distribution diagram by using a finite element based on the trend, extracting high-risk data to calculate a real-time evaluation value, and dynamically adjusting the monitoring frequency to generate an optimization strategy. By adopting the method, the reliability and economy of engineering safety monitoring can be remarkably improved.

The invention relates to the technical field of image enhancement, in particular to a protector bimetallic strip detection method based on machine vision, and the method comprises the steps: obtaining a surface image of a bimetallic strip, and carrying out the preprocessing of the surface image, and obtaining a gray image; performing threshold segmentation on the grayscale image to obtain at least one segmentation region; obtaining an evaluation coefficient of each segmented region, obtaining an adaptive cutting parameter when contrast-limited adaptive histogram equalization is carried out on each segmented region according to the evaluation coefficient of each segmented region, and carrying out image enhancement on each segmented region in the grayscale image according to the adaptive cutting parameter to obtain an enhanced grayscale image; the corrosion detection result in the bimetallic strip is obtained by using the enhanced gray level image, so that the enhanced image can reflect more effective information, and the surface corrosion detection precision of the bimetallic strip according to the enhanced image is improved.

The invention discloses a mobile terminal streetscape image real-time segmentation method based on a lightweight neural network, and relates to the technical field of image segmentation. The method comprises the following steps: firstly, carrying out 320 * 320 adjustment, Z-score standardization, adaptive histogram equalization and 3 * 3 Gaussian filtering preprocessing on an input streetscape image; then, an improved MobileNetV3 backbone network is used, and a five-scale feature map is output in combination with DropBlock regularization through eight feature extraction stages including depth separable convolution and an SE attention module; multi-scale features are fused through a U-shaped structure, and a fusion feature map is generated through up-sampling, element-by-element addition of dimension reduction low-layer features and an attention gating module; and during reasoning, outputting a segmentation mask by using a convolutional layer, Softmax and a conditional random field, and finally performing knowledge distillation, weight pruning, 8-bit quantization and TensorRT optimization. According to the invention, high-precision real-time street view segmentation is realized, the robustness is high, and the method is suitable for different devices and scenes.

The invention relates to a segmentation model based on a CNN and Transform parallel double-branch structure, and belongs to the technical field of medical data prediction. The method comprises the following steps: acquiring a CT image, and preprocessing the CT image by executing windowing processing and contrast limited adaptive histogram equalization; extracting features of lung lesions in the preprocessed CT image through a parallel double-branch structure; inputting the extracted features into a cross enhancement fusion module, and performing complementary fusion on the features through dynamic weight distribution to obtain fused features; the fusion features are input into a multi-scale context information extraction module, and lesion boundary sensitivity is enhanced through cavity convolution of different expansion rates; the encoder features and the decoder features are fused through jump connection, and a segmentation result is output after resolution is recovered based on up-sampling; and optimizing model training by adopting a weighted loss function. Accurate segmentation of the lung lesion in the pulmonary tuberculosis CT image is realized, and clearer and more accurate lesion area information can be provided.

The invention belongs to the technical field of face recognition, and particularly relates to a face recognition algorithm adaptive to illumination variation, which uses Gaussian filtering to reduce noise caused by illumination variation, convert a color image into a grey-scale image, reduce calculation complexity, perform adaptive histogram equalization, estimate global illumination conditions in the image and improve the face recognition accuracy. Calculating the main direction and intensity of illumination in the image, predicting the current illumination condition, and enabling the image brightness to adapt to different environment illumination; identifying a face region in the image, using a deep learning method to position face related points, using an LBP to extract local features insensitive to illumination variation, and using a deep convolutional neural network to extract global features; introducing a multi-illumination data set; the Euclidean distance is used for comparing the extracted feature vectors, the similarity between the extracted feature vectors and known faces in a database is recognized, according to the similarity score, threshold judgment is adopted for AQW to obtain a final recognition decision, and the method has the effect of being capable of accurately adapting to facial features under different illumination conditions in real time.

The invention relates to the technical field of image enhancement, in particular to a liver CT image enhancement method and system. The method comprises the following steps: acquiring an initial window of each connected domain; obtaining the density change degree of each initial window according to the density distribution of pixel points in each initial window on the CT image; obtaining a multi-scale image corresponding to the CT image, and obtaining the high-frequency information change degree of each initial window according to the frequency domain distribution characteristics of the initial windows corresponding to the same positions among different-scale images; obtaining a window correction coefficient of each initial window, correcting each initial window, and obtaining a correction window of each initial window; and performing adaptive histogram equalization on the images of different scales according to the correction window, and performing weighted fusion in combination with the confusion degree of the images of the corresponding scales to obtain an enhanced CT image. According to the method, the effect of influence enhancement is improved by obtaining the accurate window size during image equalization processing.

The invention discloses a carbon-coated foil coating uniformity detection method and system based on machine vision, and the method comprises the steps: S1, collecting blue light, green light, red light and near infrared spectrum images of the surface of a carbon-coated foil coating through a multi-angle camera, and generating a fused composite image through a multispectral image fusion algorithm; s2, performing adaptive histogram equalization and reflected light spot suppression processing on the fused composite image, and outputting an enhanced image; s3, coating edge features in the enhanced image are extracted, segmentation of a coating region is realized in combination with a region growing method, and a coating region mask is obtained; s4, extracting multi-scale texture features of the coating region by adopting a pre-trained ResNet-50 deep convolutional network, and constructing a coating surface feature vector; and S5, based on the extracted feature vectors and a standard template library, through a uniformity evaluation model, calculating a uniformity index of the coating, and detecting the uniformity of the coating. According to the invention, accurate detection of the coating uniformity can be realized.

The invention discloses a micro-channel aluminum flat tube appearance detection method based on image processing, and relates to the technical field of appearance detection.According to the method, more comprehensive information acquisition is carried out through the multispectral imaging technology in combination with RGB, polarized light, infrared spectrum images and ultraviolet spectrum images, and surface temperature difference changes are detected through infrared spectrums; image preprocessing is carried out through contrast-limited adaptive histogram equalization and bilateral filtering, uneven illumination is inhibited while edge details are kept, the contrast of corrosion defects is improved, details of a low-resolution area are enhanced through a super-resolution reconstruction technology, edge features of a corrosion area are further optimized in combination with a histogram in the gradient direction, and the edge features of the corrosion area are further optimized. The identification capability of small-scale defects is improved; an improved YOLOv3 defect detection algorithm is introduced, multi-scale feature extraction is adopted, and a self-attention mechanism is introduced, so that the model pays more attention to a tiny corrosion area, and meanwhile, the detection precision of a small target is enhanced in combination with focus loss and regression loss.

The invention relates to the technical field of image processing, and discloses an intelligent weight monitoring method, system and device for bred pigs and a storage medium. The method comprises the following steps: acquiring an original data set through multi-angle image acquisition; carrying out image processing by utilizing adaptive histogram equalization and Gaussian filtering; obtaining body shape parameters by adopting contour-skeleton feature extraction and curvature analysis; performing multi-source fusion processing in combination with the data of the Internet of Things; body weight modeling is carried out based on a hierarchy-increment prediction model and an attention mechanism; and generating a feeding scheme through reinforcement learning. According to the invention, the technical problems of insufficient data fusion and low prediction precision in the existing breeding pig weight monitoring method are solved.

The invention discloses an infrared image enhancement method based on scene segmentation. The method comprises the following steps: (1) inputting an original infrared image into an infrared image scene segmentation network based on edge feature guidance to obtain a scene soft segmentation result; (2) decomposing the infrared image into a background layer image and a detail layer image by adopting bilateral filtering; (3) performing adaptive enhancement on the detail layer image based on characteristics of different scenes; (4) enhancing the background layer image by adopting contrast limited adaptive histogram equalization; and (5) carrying out weighted fusion on the enhanced detail layer image and the background layer image. According to the infrared image enhancement method based on scene segmentation, the semantic segmentation algorithm is introduced, the infrared image is divided into different scene areas, and the adaptive infrared image detail enhancement parameters are set for the detail information and noise features of each scene area, so that the quality of the infrared image can be effectively improved.

The invention discloses a multi-scale feature detail enhanced lithium battery surface defect detection method and system, and relates to the technical field of defect detection, and the method comprises the steps: carrying out the lithium battery region segmentation of a to-be-detected lithium battery surface image, and obtaining a lithium battery region image; performing pyramid decomposition and adaptive contrast limit adaptive histogram equalization based on the lithium battery region image through a multi-scale adaptive enhancement module, and outputting a plurality of scale feature maps; respectively inputting each scale feature map into a comprehensive enhancement module for detail enhancement, and determining a corresponding detail feature map; performing up-sampling fusion on each detail feature map by adopting a feature fusion module to generate an enhanced lithium battery region image; and performing defect detection on the enhanced lithium battery area image based on the target detection model, and outputting a detection result. Based on the scheme, the defect detectability is improved through enhancement treatment, and the lithium battery surface defect detection reliability can be improved.

The invention discloses a cancer image contrast enhancement method and system based on deep learning, belongs to the technical field of medical image processing, and adopts an adaptive histogram equalization algorithm for preprocessing to enhance the contrast of a local region. Texture, shape and edge features are extracted through multi-scale analysis, and feature descriptors are constructed. And based on the feature descriptor, segmenting the image by using a region growing segmentation algorithm to obtain a focus region boundary. A self-adaptive contrast enhancement technology is adopted for the segmented focus area, and the focus display effect is highlighted. In consideration of noise interference, a denoising method based on wavelet transform is introduced, noise is suppressed in a high-frequency sub-band, and details are reserved in a low-frequency sub-band. Through the organic combination of the above technical schemes, the contrast and definition of the cancer image can be effectively enhanced, the focus area is highlighted, and noise interference is inhibited at the same time.

The invention provides an auxiliary pathological analysis method for a microscope system, and belongs to the technical field of pathological analysis, and the method comprises the steps: firstly carrying out the gray normalization, denoising and other preprocessing of an obtained pathological section image, and carrying out the illumination equalization processing through a region adaptive histogram equalization algorithm; then, color features of the image are extracted in an LAB color space, a color representation center is calculated, a feature mapping matrix is established, and meanwhile texture features are extracted and fused with the color features; and carrying out dimension reduction on the fused feature vector by using a principal component analysis algorithm, and reserving important feature information. And then, establishing a convolutional neural network model, inputting the feature vectors after dimension reduction, performing training and cross validation by using a labeling data set, and improving the model performance through iterative optimization. According to the invention, the technical problem that the diagnosis and judgment of a doctor are seriously influenced due to inconsistent color expression caused by the difference of shooting conditions of pathological section images observed by a microscope is solved.

The invention belongs to the technical field of image processing, and particularly relates to a monitoring image defogging enhancement method in a high-dust environment of a mineral separation site, which comprises the following steps of: performing smooth denoising on an original image by using weighted guided filtering, and obtaining global atmospheric light through a quadtree subdivision method; constructing a same-color heterogeneous discrimination index combining a spectrum similarity factor and a texture confidence factor for distinguishing dust and ore with similar colors; calculating a pixel-level dynamic defogging coefficient based on the discriminant index, and obtaining an adaptive transmissivity in combination with dark channel prior; and finally, restoring the image by using an atmospheric scattering model and carrying out contrast-limited adaptive histogram equalization processing. According to the method, the problem of misjudgment caused by the fact that the colors of the ore and the dust on the ore dressing site are similar is effectively solved, powerful defogging of the dust area and detail reservation of the ore area are achieved, and the definition and the contrast ratio of the monitoring image are improved.

The invention relates to the technical field of underwater image enhancement, in particular to an underwater image enhancement method based on balance, correction and deblurring and application. The balance module is cascaded by color balance and illumination balance, and the color balance performs compensation balance on RGB (Red Green Blue) color channels of the underwater image based on a white balance algorithm of red channel compensation; according to illumination balance, an improved contrast-limited adaptive histogram enhancement method is used for enhancing illumination of a low-illumination area of an underwater image, and meanwhile illumination of a high-illumination area is restrained, so that illumination balance is achieved. In the correction module, a color and illumination composite correction method combining color constancy and an illumination diagram is provided so as to correct the color and illumination of the underwater image at the same time. And in the deblurring module, the image is deblurred by adopting a multi-scale anti-sharpening mask, so that the definition of the underwater image is improved.

The present invention discloses an AI-based automated liver cancer tissue recognition and diagnosis system, which relates to the field of medical impact processing technology, including a data upload terminal, an image processing terminal, a liver cancer tissue algorithm recognition terminal, a liver cancer automatic diagnosis terminal and a user interface terminal; the data upload terminal is responsible for uploading the patient's liver image data to the cloud or local analysis platform of the system through a secure transmission channel for processing; the image processing terminal is responsible for preprocessing the acquired image data, including denoising, contrast enhancement and artifact correction operations; using a filtering algorithm to remove noise in the image; and using adaptive histogram equalization to increase the image contrast. The present invention can automatically extract the subtle features of liver cancer lesions and improve the recognition accuracy of different lesion morphologies; compared with the traditional subjective judgment that relies on the doctor's experience, the system shows higher sensitivity in detecting small lesions and early liver cancer, reducing the probability of misdiagnosis and missed diagnosis.

The invention relates to the field of hydrological monitoring, in particular to a hydrological monitoring early warning method and system based on an unmanned aerial vehicle. The method comprises the following steps: acquiring a panoramic continuous aerial image of a to-be-monitored area; adaptive histogram equalization and image time delay interpolation optimization are carried out on the panoramic continuous aerial image of the to-be-monitored area to construct a time delay optimized continuous image; performing water body landform semantic analysis on the time delay optimized continuous image, performing real-time three-dimensional landform modeling, and constructing a real-time three-dimensional landform model; performing continuous frame optical flow tracking on the time delay optimized continuous image, performing inter-frame optical flow motion vector calculation, and generating a water level fluctuation curve and surface ponding area change characteristics; and carrying out instant hydrological state rendering on the real-time three-dimensional terrain model according to the water level fluctuation curve and the surface ponding area change characteristics so as to construct a dynamic hydrological state evolution model. According to the invention, accurate and efficient hydrological monitoring and early warning are realized, and the safety of the hydrological basin is improved.

The invention relates to the technical field of computer vision, and discloses a jade defect intelligent detection method and system based on machine vision and deep learning, and the method comprises the following steps: S1, based on a high-resolution industrial camera and a laser three-dimensional scanner, adopting a multi-mode synchronous collection strategy, and rotating a jade sample through a precise motion control system, a jade surface high-resolution two-dimensional color image and high-precision three-dimensional point cloud data are respectively obtained, and a jade multi-mode original data set is generated. A high-resolution two-dimensional color image and high-precision three-dimensional point cloud data are integrated through a multi-modal synchronous acquisition strategy, and multi-dimensional feature expression under unified coordinates is constructed, so that the limitation of a single data source is effectively overcome; an image registration algorithm and a feature pyramid network are combined with a point cloud network to perform multi-modal feature fusion, complementarity of color texture and geometric morphology information is enhanced, and image quality is optimized based on adaptive histogram equalization and non-local mean filtering.

The invention provides a real-time processing method for endoscopic image blood vessel enhancement and microcirculation evaluation, which comprises the following steps of: processing an original endoscopic image, and locally enhancing a brightness channel through a contrast-limited adaptive histogram equalization technology; self-adaptive frequency domain-space domain decomposition is realized based on local texture complexity analysis; performing multi-scale Hessian matrix blood vessel detection on the low-frequency component; applying a directional Gabor filter bank to the high-frequency component; spatial-temporal feature fusion is realized through multi-resolution pyramid optical flow calculation; synchronously completing blood vessel probability prediction, blood vessel diameter estimation and blood flow direction prediction by using a lightweight multi-task deep learning network; the blood flow velocity is analyzed and calculated based on a speckle mode, the perfusion density is subjected to accelerated statistics through an integrogram, and vascular morphological parameters are extracted by adopting an improved skeleton algorithm. According to the invention, an enhanced blood vessel visualization effect and a real-time microcirculation quantitative evaluation function can be provided, and the overall improvement of the endoscope image processing quality and efficiency is realized.

The invention relates to the technical field of image processing, and discloses a gynecological cell morphology intelligent identification method and system based on machine vision, and the method comprises the steps: collecting an original image of gynecological cells, carrying out the color normalization processing of the collected original image, carrying out the denoising processing through employing a wavelet threshold value, and carrying out the image enhancement through employing adaptive histogram equalization; according to the method, color consistency errors of different batches of dyed images are reduced, noise introduced in the image acquisition process is eliminated, cell structure details are reserved, and the accuracy of diagnosis is improved. And then image enhancement is carried out, the contrast ratio of cell nucleuses and cytoplasm is improved, so that the characteristics in the cells are more obvious, segmentation can be more accurately completed during subsequent image segmentation, the lesion level of the gynecological cells can be accurately identified, and the accuracy of an identification result is ensured.

The invention relates to a balanced color perception enhancement method for rail transit target detection, and belongs to the technical field of rail transit and computer vision. According to the method, brightness channel adaptive histogram equalization enhancement CLAHE-LC, a multi-segment tone channel mask mechanism MSHCM-M and a three-stage hybrid mechanism non-maximum suppression TSLSH-NMS technology are combined, so that the problems of complex illumination, multi-target shielding, color interference and the like in a rail transit scene are solved, and the accuracy and recall rate of target detection are improved. The method can be seamlessly integrated into an existing deep learning target detection framework, is compatible with multi-label, multi-category and multi-scale features, and remarkably improves the precision and recall rate of target detection in a rail transit scene. Experimental verification shows that the real-time performance is guaranteed, meanwhile, the detection performance is obviously improved compared with a traditional scheme, and the method is suitable for being applied to an actual rail transit safety monitoring and intelligent maintenance system.

The invention relates to the field of tea disease detection, in particular to a tea disease detection method based on a lightweight YOLOv8 model. According to the technical scheme, the method comprises the steps of collecting various types of tea disease images; the method comprises the following steps: preprocessing multiple types of collected tea disease images, including illumination normalization, geometric enhancement, illumination enhancement, adaptive histogram equalization, a multi-scale Retinex algorithm, dynamic Gamma correction and image noise removal, and marking scab regions in different types of tea disease images to obtain a training data set; a lightweight YOLOv8 model is constructed; using the training data set to train the lightweight YOLOv8 model; and inputting a collected tea disease image into the trained lightweight YOLOv8 model, and outputting a tea disease type through the trained lightweight YOLOv8 model. The method is suitable for tea disease detection.

The invention discloses an SLAM front-end optimization method based on brightness grading and gradient constraint, and relates to the technical field of computer vision and vision SLAM. In order to solve the defects that adaptive grading enhancement and parameterization control based on illumination types are lacked in the prior art, and feature point quality, spatial distribution uniformity and front-end real-time performance are difficult to consider in a high-frequency input scene, the invention provides a comprehensive brightness grading and feature optimization scheme. The method comprises the following steps: firstly, calculating the average brightness of an input image, dividing the image into a dark light type, a normal type and an overexposure type, and respectively adopting gamma correction, contrast limited adaptive histogram equalization and inversion enhancement strategies for different types to realize illumination adaptive enhancement; then screening high-quality feature points with significant local curvature changes; and updating the detection area. According to the method, the feature stability, the matching precision and the real-time performance of the SLAM front end in a complex indoor environment are remarkably improved, and the method is suitable for a self-localization and mapping system.

The invention discloses an underwater image enhancement method and system. The method comprises the following steps: performing L-channel contrast enhancement on an original underwater image to obtain a local enhanced image; performing color recovery on the original underwater image by using a contrast-limited adaptive histogram equalization method to obtain an equalized image; performing linear weighted fusion on the locally enhanced image and the equalized image to obtain a fused image; performing color correction on a # imgabs0 # channel and a # imgabs1 # channel on the fused image to obtain a corrected image; performing depth prediction on the fused image through a monocular depth estimation method to obtain a depth guide map; carrying out RGB channel information expansion on the fused image to obtain a detail image; and carrying out image fusion on the corrected image, the depth guide image and the detail image to obtain a final enhanced image. According to the method, the coordination and fusion of local contrast enhancement and global color recovery are realized, so that the visual quality of the enhanced image is remarkably improved.

The invention discloses a pavement crack image segmentation method based on a convolution block attention and diffusion model, and belongs to the technical field of computer vision deep learning application. The problem of excessive enhancement caused by traditional histogram equalization is solved through an adaptive histogram equalization algorithm, the contrast of a crack region is remarkably improved while texture details are reserved, and a higher-quality data basis is provided for subsequent segmentation; a segmentation framework combining a convolution block attention mechanism and a diffusion generation model is adopted, crack features are accurately positioned through a space-channel double attention mechanism, interference of road background noise is effectively inhibited, and the crack segmentation accuracy is improved; a feature-level joint training mechanism is adopted, convolution attention enhancement of an FEF module is matched, multi-dimensional feature complementation is achieved, multi-scale features are effectively combined, space positioning of a crack area is enhanced, an end-to-end model is constructed, and convenience of image segmentation is enhanced.

The invention discloses an unmanned aerial vehicle small target detection method based on Mama feature fusion, and relates to the technical field of computer vision, and the method comprises the steps: multi-modal data collection, synchronous collection of images and point cloud data through three types of sensors, and coverage of multi-scene and environment conditions; image preprocessing adopts improved bilateral filtering, adaptive histogram equalization and point cloud downsampling to unify a coordinate system; in the multi-scale feature extraction, five-level scale features are output through an improved CSPDarknet network, and the five-level scale features are enhanced through an SENet attention module; in the Mama feature fusion, multi-modal and multi-scale features are processed through a three-stage unit; generating a candidate frame by a dynamic anchor frame, screening according to IOU, and improving YOLOHead to realize classification and positioning; the dynamic optimization of the detection result finely adjusts parameters through on-line distillation. According to the method, the precision, the real-time performance and the anti-interference capability of small target detection of the unmanned aerial vehicle are improved, and reliable technical support is provided for low-altitude security, exploration and other scenes.