7416 results about "Image quality" patented technology

A pattern which does not appear at a flat portion in normal binarization processing is set as a code pattern, and a code formed from this pattern is attached. At this time, code attachment with little degradation in image quality is implemented by selecting an unnoticeable pattern.

The invention discloses an image enhancement method and system in a complex coal mine environment, and relates to the technical field of image processing, and the method comprises the steps: carrying out the preprocessing of a collected coal mine image of a target region, and dividing the coal mine image into different semantic regions, including a bright region, a dark region and a dust shielding region, through a deep learning semantic segmentation model; according to semantic region characteristics, a differentiation enhancement strategy is made; a traditional Retinex model is improved, non-local mean filtering is introduced, and an illumination component and a reflection component are decomposed through pixel similarity matching. According to the method, the image is divided into the bright area, the dark area and the dust shielding area through the deep learning semantic segmentation model, differential enhancement strategies are formulated according to different area characteristics, detail distortion caused by global adjustment is avoided, local contrast suppression is adopted in the bright area, illumination compensation is enhanced in the dark area, and the image quality is improved. Noise diffusion of the dust shielding area is inhibited through edge preservation smoothing, the image quality of each area is remarkably improved, and it is ensured that image details in a complex coal mine environment are clear and visible.

Systems and methods are disclosed for classifying objects using electro-optical and synthetic aperture radar images through multi-modal feature alignment and fusion. A computing system acquires and preprocesses image data, then aligns features across modalities using a multi-modal alignment engine. A cross-modal attention fusion network extracts and integrates complementary information using transformer-based attention mechanisms. A modality-specific feature extraction framework processes EO and SAR images through specialized branches, ensuring optimal feature representation. An adaptive fusion decision system dynamically determines the best fusion strategy based on image quality and confidence scores. A self-supervised consistency controller enforces alignment between EO and SAR features using contrastive learning. The fused representations are processed by a neural network to generate object classifications. This system improves accuracy and robustness in environments where one modality may be degraded or missing, enhancing applications such as remote sensing, surveillance, and autonomous navigation.

The invention discloses a manipulator grabbing method based on deep learning target detection and image segmentation, and relates to the technical field of artificial intelligence and robotics.The manipulator grabbing method comprises the following steps that a scene image to be processed is collected, the image quality is improved through the multi-light-source fusion image enhancement technology, and recognition errors caused by uneven illumination are reduced; and inputting the enhanced image to a pre-trained deep learning model, executing a target detection task, and outputting an initial bounding box and a category label of the target object. According to the method, through multi-light-source image enhancement and high-precision image segmentation, the accuracy of target recognition and contour extraction is remarkably improved, and the capture failure rate caused by image misjudgment is reduced. And meanwhile, geometric consistency verification and a multi-factor grabbing scoring mechanism are introduced, dynamic screening and collision pre-detection are conducted on the paths, the grabbing stability and safety of the mechanical arm in the complex environment are effectively guaranteed, and the intelligence and robustness of the whole system are remarkably improved.

The invention discloses a face image enhancement and recognition method in a low-light environment, and relates to the technical field of face recognition. Firstly, multi-scale Retinex enhancement is carried out on a low-illumination face image, brightness and details are improved, then LBP and HOG features of the enhanced image are extracted, weight fusion is dynamically adjusted according to illumination, then space and channel weighted optimization is carried out on the fused features by using a residual attention module, finally, optimized features are input into a pre-training model to extract vectors, and finally, the pre-training model is used for pre-training. And through cosine similarity matching identification, face processing and identity determination under low illumination are completed. According to the method, the problem of low-illumination face recognition is solved, the image quality is improved through multi-scale enhancement, feature extraction is enhanced through dynamic fusion and an attention mechanism, and the recognition accuracy and stability are improved; in addition, living body detection is added to guarantee safety, and the method is suitable for multiple scenes such as security and protection, finance and the like.

The invention relates to the technical field of image data processing, in particular to a cable surface defect detection method and system based on machine vision. The method comprises the following steps: acquiring a surface image of a cable, and converting the surface image into a grayscale image; determining the local complexity of each pixel point; obtaining a plurality of areas of the grayscale image, performing complexity determination, and dividing each area to obtain a plurality of windows of each area; determining a contrast limit threshold value of each window; performing image enhancement by using a CLAHE algorithm to obtain an enhanced grayscale image; and carrying out cable surface defect detection on the enhanced grayscale image by using a defect detection algorithm. According to the method, the CLAHE parameters are adaptively adjusted based on local complexity, the window size and the contrast threshold are dynamically determined in combination with gray and gradient information, discontinuity is corrected and eliminated through boundary similarity, and the quality and reliability of a cable defect detection image are improved.

The invention provides a multi-layer optical film detection method, device and equipment and a storage medium, and the method comprises the steps: obtaining a polarization phase difference signal of transmission light and reflected light through multi-angle incidence irradiation of a multi-band programmable laser source, and recording the polarization ellipticity parameter change; then combining confocal microscopic interference and polarization analysis to determine a space coordinate and a stress distribution diagram of a stress abnormal region; continuously irradiating in a photo-thermal coupling environment through a broadband light source, and monitoring interference signals, polarization states and temperature changes to obtain dynamic evolution characteristics; and finally, performing a simulation test on the multi-layer optical film in a system environment with preset optical axis arrangement and light beam energy distribution, obtaining an in-situ polarization state and interference pattern data, and evaluating the optical performance of the film layer. According to the invention, potential defects can be positioned under high resolution, and the reliability and imaging quality of the film layer under complex application conditions can be accurately evaluated.

Provided are a quantification method for microscopic cracks inside 3D printed concrete and a system thereof. Before measuring the microscopic crack images according to relevant standards, firstly, a denoising network, improved attention-guided denoising neural network (IADNet) is adopted. IADNet can extract features from microscopic crack images from different perspectives, perceive noise from multiple levels, and perform denoising processing, greatly improving image quality and enhancing texture details, which is beneficial for training segmentation networks. The combination of IADNet and semantic segmentation algorithm has the ability to finely recognize image information, quantify microscopic crack recognition, overcome the shortcomings of measurement and analysis of microscopic cracks inside 3D printed concrete, and improve construction efficiency and quality.

The invention discloses a light guide plate defect detection method and system based on a neural network, and particularly relates to the technical field of machine vision detection, and the method comprises the following steps: aiming at the problem of image instability of a light guide plate in a dynamic transmission or rotation process, continuously collecting an image sequence and extracting time domain features; and performing interference judgment in combination with the inter-frame consistency prediction coefficient and a first threshold to realize accurate identification of the abnormal image frame. For an abnormal image frame, further correcting the recognition credibility of the abnormal image frame by adopting a confidence adjustment and fusion mode, and meanwhile, introducing a frequency domain transformation and image enhancement strategy to compensate detail loss caused by motion blur; according to the method, inter-frame consistency analysis, confidence fusion regulation and control and frequency domain fuzzy recognition and compensation mechanisms are introduced, abnormal judgment and image quality restoration of the light guide plate image in the dynamic scene are realized, the recognition accuracy and stability of the neural network model on the defect type, position and confidence are improved, and the false detection and omission ratio is effectively reduced.

The invention relates to the technical field of unmanned aerial vehicle flight attitude control, in particular to an unmanned aerial vehicle flight attitude adjustment method and system based on reinforcement learning. Comprising the following steps: constructing a dynamic environment grid map based on a digital map and a real-time semantic segmentation result, and generating an initial track by introducing a space-time constraint fast search random tree algorithm; collecting flight state information, environment perception information and image definition indexes of the unmanned aerial vehicle, inputting the flight state information, the environment perception information and the image definition indexes into a space-time attention encoder, and generating a semantic-fused state tensor; performing importance distribution on the state tensor through an information entropy weighting mechanism to obtain a weighted state vector; inputting into a Meta-SAC model with meta-learning ability, and outputting a target flight reference pose and an LQR controller dynamic gain coefficient; the control module drives the LQR controller to generate a flight control instruction based on the information; and a reward function is constructed based on the image quality and the energy consumption efficiency, and a reinforcement learning strategy is fed back in real time.

The invention discloses a package defect elimination method based on visual inspection, and relates to the technical field of image processing and computer vision, and the method comprises the following steps: setting a polarized light source with a fixed angle on a package production line, and uniformly illuminating a target package area to enable the incident light polarization direction and the package surface reflection direction to form a deviation angle; an industrial camera provided with a polarization filter is used for carrying out image collection on a packaging target area, and the polarization direction of the filter is perpendicular to the polarization direction of a light source so as to restrain reflection interference to the maximum extent. Reflection interference is eliminated through polarized light, and the image quality is improved; by combining a neural network and feature fusion scoring, the defect recognition accuracy and adaptability are improved; a closed-loop elimination control and data tracing mechanism is introduced, the elimination precision is guaranteed, quality responsibility investigation is supported, and the intelligence and reliability of the packaging defect detection and elimination system are integrally improved.

The invention discloses an AOI optical scheme automatic optimization method based on reinforcement learning, and belongs to the technical field of automatic optical detection. According to the method, a reinforcement learning technology is applied to automatic adjustment of optical schemes in a machine vision system, wafer images under different optical schemes are collected by using an AOI system, and scoring is performed by using a plurality of semantic segmentation models; constructing and training a non-reference image quality evaluation network, evaluating the optical imaging quality in real time, and designing a reward function on the basis; by using the Actor-Critic algorithm, the intelligent agent can autonomously learn an optimal parameter adjustment strategy, quickly adapt to different scenes, realize dynamic optimization of light source parameters and remarkably improve image quality, and an efficient and intelligent solution is provided for efficient application of a machine vision system in a complex industrial environment.

The invention relates to the technical field of computer vision, and discloses a thyroid ultrasonic robot automatic scanning method, device and equipment based on RGB images and depth information and a medium, and the method comprises the steps: obtaining image information and depth information, coding the depth information, fusing and recognizing a target scanning area, determining an initial scanning point and an initial scanning direction, controlling the scanning probe to scan and collect a real-time scanning image, analyzing the real-time scanning image to recognize a preset target and an artifact area, adjusting a scanning posture and a scanning path based on a recognition result, monitoring a continuous existence state of the preset target, and stopping scanning when the preset target is not recognized continuously. The target area is identified by fusing the multi-modal image information, the scanning posture and path are dynamically adjusted in combination with real-time image analysis, scanning termination is intelligently controlled according to the target detection result, the positioning accuracy, image quality and standardization level of ultrasonic scanning are improved, and the method is suitable for automatic ultrasonic imaging of thyroid and superficial organs.

The invention provides an unmanned aerial vehicle multi-channel image transmission optimization system based on link quality perception so as to improve the transmission stability and the image quality guarantee capability of image data in a complex wireless environment. The prediction module constructs a time sequence model based on the historical link quality index of the wireless channel, and outputs a future link stability prediction value; the modeling module analyzes texture change and the like between the image frames, generates an image frame evolution vector, and calculates the aging sensitivity and reconstruction importance score of the image frames according to the image frame evolution vector; the decision-making module fuses the information and generates an optimal matching relation between the image frame and the channel and a transmission priority parameter; the grouping module performs image frame clustering according to the similarity between the priority parameters and the evolution vectors, constructs compression groups and generates corresponding compression configuration files and compression image data; the transmission module schedules the compressed image data to a corresponding wireless channel for transmission; the system can be widely applied to an unmanned aerial vehicle image transmission task with relatively high requirements on real-time performance and image quality in a dynamic environment.

The invention discloses a global nerve drawing method and system for a multivariate mixed representation scene, and belongs to the technical field of computer graphics, and the method comprises a scene expression module which is used for loading multivariate mixed scene data; the geometric module is used for generating a multi-channel geometric buffer based on the multivariate mixed scene data; the illumination module comprises at least one neural network sub-module and is used for carrying out illumination calculation based on multi-channel geometric buffering; the scheduling and synchronizing module is used for coordinating cross-module data streams through a rendering graph structure and executing resource merging and task scheduling between the geometric module and the illumination module based on a unified request structure; and the merging output module is used for fusing neural network illumination output and traditional rendering output to generate a final frame image so as to complete global neural rendering. According to the invention, multi-element scene data is integrated through modular design, efficient collaboration of traditional and neural rendering is realized by using a unified interface and a scheduling structure, and the image quality and the system flexibility are improved.

The invention belongs to the technical field of computer data processing, and provides a biodiversity inversion method based on multi-source remote sensing image fusion. Comprising the steps of remote sensing image data acquisition, image preprocessing, image fusion processing, multispectral resolution image data generation, spectral feature extraction, final frequency feature extraction, feature integration and target ecological variable prediction. According to the invention, through wave basis adaptive selection and multi-scale wavelet decomposition, spectrum fidelity and space structure maintenance are considered, differential fusion of high and low frequency components under different scales is realized, and multi-source data complementarity and fusion image quality are improved; through spectral resolution refinement processing and multi-bandwidth scale simulation, the limitation of single resolution is broken through, and the capability of capturing complex spectral features of vegetation is enhanced; the spatial correlation is enhanced through spatial neighborhood feature fusion; and through an ecological variable inversion estimation model, multi-index synchronous prediction is realized, and the universality of the model is improved.

The invention discloses a packaging box printed matter printing quality detection method, which comprises the following steps of: acquiring a high-resolution printing image through an image acquisition module, preprocessing and segmenting the printing image by an image processing module, optimizing image quality, classifying the preprocessed image by an image classification module according to characters, patterns and color blocks, and detecting the printing quality of a packaging box printed matter. The defect detection module performs multi-dimensional detection on the classified preprocessed images based on image parameter data, the quality analysis module quantitatively evaluates quality scores of characters, patterns and color blocks based on detection results, and the computer terminal generates a comprehensive score through weighting calculation and automatically generates a detection report; according to the packaging box printed matter printing quality detection method, complex defects such as font errors, stroke breakage, pattern deviation, color deviation and stains are effectively recognized through multi-dimensional defects, the manual reinspection cost is reduced through full-process automatic processing, and the quality of printed matter is visually reflected through a quantitative evaluation system.

The invention relates to the technical field of medical image processing, in particular to an artificial intelligence assisted intraoperative imaging method, which comprises the following steps: S1, preprocessing a multi-modal medical image, segmenting and recognizing an anatomical structure by a deep learning model according to the preprocessed image, measuring anatomical parameters based on a segmentation and recognition result, and generating an operation planning path by artificial intelligence according to the anatomical parameters; s2, collecting a C-shaped arm perspective image stream in real time, dynamically tracking space coordinates of a surgical instrument, comparing the position of the instrument with a surgical planned path, calculating offset, and when the offset is greater than an offset threshold, outputting correction guidance through an AR superposition layer; and S3, monitoring an image quality index in real time, dynamically adjusting exposure parameters through a reinforcement learning model, and when a metal implant is detected, switching a dual-energy-spectrum mode and executing an artifact suppression algorithm. According to the method, preoperative precise planning and intraoperative assistance are realized through artificial intelligence, the problems of poor image quality and high radiation risk are solved through technical optimization, and the method has important clinical application value.

The invention discloses a power transmission image compression quality evaluation method based on multichannel fusion, belongs to the technical field of smart power grids, and solves the problem of how to enhance adaptive evaluation performance of image quality in scenes with different compression ratios. A multichannel feature processing module introduces learnable offset to realize dynamic alignment and effective fusion of multichannel features; the multi-scale attention fusion module is combined with multi-scale feature extraction and an attention mechanism, learnable position codes are introduced, and the sensitivity to local structure changes is enhanced while the global perception ability is ensured; the self-adaptive quality evaluation module captures different scale features and local information by constructing residual connection and a multi-branch structure, performs weighted fusion on multi-channel fusion features in combination with a dynamic weight generation mechanism, and realizes accurate judgment of the quality of a complex scene image under different compression levels; while the image compression distortion perception capability is improved, the adaptive evaluation performance of the image quality in scenes with different compression ratios is effectively enhanced.

The invention relates to the technical field of unmanned aerial vehicle control, in particular to an unmanned aerial vehicle shooting system control method based on adaptive optimization. A multi-modal semantic map fusing task description, equipment types, geographic positions, historical data, illumination and weather information and image optical flow features is constructed, and an equipment space distribution probability, a shooting difficulty level and key route nodes are obtained by adopting graph neural network reasoning. On the basis, a control strategy candidate set is generated, and optimal shooting parameter configuration is screened out in combination with a Bayesian optimization algorithm. The unmanned aerial vehicle collects multi-source information in real time in the flight process, dynamic fusion is conducted through an attention mechanism, the combined control module is driven to synchronously adjust the attitude of a holder and camera parameters, and accurate imaging control in a complex scene is achieved. And when the recognition confidence is low, triggering a supplementary shooting control mechanism based on the semantic map, and performing local fine tuning to improve the image quality. And the system also continuously updates the control strategy through transfer learning, so that the adaptability to a new task environment is enhanced.

The invention discloses a weld quality intelligent diagnosis system based on deep learning, and relates to the technical field of welding quality detection, and the weld quality intelligent diagnosis system comprises an image quality evaluation module, a feature alignment module, a deviation detection module, a path reconstruction module, a prior enhancement module and a defect identification module, identifying an area of which the signal-to-noise ratio is lower than a preset threshold value, and constructing a noise interference distribution diagram; and the feature alignment module executes a deformable convolution feature alignment operation with a confidence factor adjustment mechanism based on the noise interference distribution diagram to generate an initial space mapping result. Through mechanisms such as image quality perception, robust alignment, deviation detection, self-adaptive reconstruction and prior enhancement, a closed-loop weld joint intelligent diagnosis process is constructed, false alignment errors are effectively inhibited, the multi-modal fusion stability and the defect recognition precision are improved, and the reliability and the intelligent level of the system under complex working conditions are enhanced.

The invention relates to an image frame insertion method and device in virtual shooting and a storage medium. The method comprises the following steps: acquiring at least two frames of reference images generated by a rendering engine; obtaining motion information and depth information of at least two frames of reference images generated by a rendering engine; and based on the motion information and the depth information of the at least two frames of reference images, determining a pixel value of a frame insertion image between the at least two frames of reference images, the frame insertion image and the at least two frames of reference images being used for sending to a display screen for display, and the display screen serving as a shooting background in virtual shooting. According to the embodiment of the invention, on-screen content can be smoothly played without improving the performance of a rendering end, the picture quality of a picture is remarkably improved, the data output capability of a rendering engine is fully played, redundant calculation is avoided, the computing power is saved, the remarkable frame rate improvement can be realized at relatively low cost, and the user experience is improved. The method meets the requirements of a virtual shooting scene, and has good stability and robustness.

The invention discloses a robot vision-inertia SLAM method and device and a medium, and belongs to the technical field of computer vision and robot navigation. The method comprises the following steps: synchronously acquiring images and inertial data through a robot binocular camera and an IMU, performing feature enhancement on an original image in combination with a pre-trained deep learning model, introducing an adaptive brightness compensation mechanism and designing an image light supplementing module based on a generative adversarial network (GAN), recovering low-illumination image details, and improving the image quality of a low-illumination area; an entropy-based adaptive dynamic interference rejection algorithm is provided, and dynamic interference feature points are rejected in combination with IMU (Inertial Measurement Unit) data; a low-rank approximate improved graph optimization algorithm is adopted, and global map construction and pose optimization are accelerated; and through entropy-based nonlinear dynamic smoothing coefficient adjustment, the track stability is improved. According to the method, the positioning precision and robustness in a low-light environment are improved, the calculation efficiency is improved, and dynamic interference is effectively resisted.

The invention discloses an image enhancement method for brightness gain self-adaptive regulation and control under low illumination, and relates to the technical field of image data processing. The image enhancement method for brightness gain adaptive regulation and control under low illumination comprises the following steps: S1, acquiring original image data and auxiliary processing data under a low illumination environment for preprocessing, and constructing an image processing database after storing the original image data and the auxiliary processing data; s2, performing illumination distribution optimization and brightness correction operation through brightness deviation analysis, and outputting an illumination enhancement component; s3, performing de-noising and detail enhancement processing according to a noise color deviation result, and outputting a corrected reflectivity component; s4, through image quality enhancement analysis, using a loss function to guide a double U-Net architecture to carry out detail reconstruction and color optimization on the image; and S5, based on the quality feedback and the difficult sample, adjusting processing parameters and driving algorithm closed-loop optimization. The problems of insufficient image brightness, noise interference, color distortion and detail missing multiple degradation in a low-light environment are solved.

The invention relates to the technical field of image processing, in particular to an image segmentation system for medical diagnosis, which comprises a regional characteristic analysis module, a modal selection optimization module, an image local enhancement module, an edge characteristic analysis module and a boundary optimization adjustment module. According to the medical image segmentation method, the gray level distribution, the texture density and the structure contour of the tissue structure in the image layer are analyzed, the system is allowed to accurately measure the characteristic deviation between different modes, the accuracy of medical image segmentation is improved, especially on the processing of the complex tissue structure, the adjacent tissues with different properties can be better analyzed and distinguished, and the medical image segmentation accuracy is improved. Real-time evaluation and adjustment of local contrast enable details of the image to be clearer, image quality is improved, accuracy of edge recognition is improved, by analyzing edge curvature and morphological characteristics, the system can optimize edge trend and reconstruct a boundary path, overall performance of image segmentation is further improved, and image segmentation efficiency is improved. Therefore, the treatment and prognosis effects of the patient are ensured.

The invention relates to the technical field of image analysis, in particular to a camellia oleifera disease and insect pest recognition system based on an image recognition technology, which realizes the structured expression of an image sequence by combining the correlation characteristics of an image time sequence and a scab evolution trajectory, forms a time axis index by using shooting time and cleans redundant images. The method effectively improves the continuity and reliability of input data, comprehensively judges the morphological dynamic state of disease spots and accurately depicts the evolutionary states of disease spot expansion, color change and the like through the comparison mode of edge texture change and main color center offset paths, remarkably enhances the staged judgment capability of disease and pest development, improves the image semantic understanding capability, and improves the image quality. Through a multi-dimension matching strategy of a closed region, a color shift direction, area jump and the like, a calibration coding sequence with traceability and staged grading capability is constructed, and the discrimination sensitivity of the system to the early stage, the middle stage and the diffusion stage of diseases is greatly improved.

The invention discloses a microscopic image seamless splicing and enhanced reconstruction method based on deep learning, and the method comprises the following steps: S1, collecting a plurality of original images with overlapped regions, and recording the spatial position information and imaging parameters of the original images; s2, preprocessing the original image to generate a standardized image sequence; s3, inputting the standardized image into a structure perception feature extraction network, and extracting a feature map fusing textures and structures; s4, inputting the feature image and the original image into an image registration module; s5, inputting the registration image into the boundary attention splicing network; s6, inputting the seamless image into the residual hierarchy reconstruction network, and enhancing image details through hole convolution and multi-scale branches; s7, image quality evaluation is executed, and the structural similarity, the signal-to-noise ratio and the edge retention rate are calculated; and S8, constructing a training set and carrying out end-to-end training optimization based on a joint loss function. According to the method, a multi-module deep network is fused, and seamless splicing and high-quality enhanced reconstruction of microscopic images are realized.

The invention relates to a plastic film defect detection device in the field of detection equipment, which is provided with a fixed clamping assembly and a movable clamping assembly to tension intermittently conveyed plastic films section by section instead of a floating roller tensioning mode adopted by a traditional film detection device, so that film vibration and motion artifacts during continuous conveying and shooting of the films are reduced, and the detection efficiency is improved. The shot plastic film has better flatness, the quality of the shot image is improved, and the accuracy of defect detection is further improved; meanwhile, the two ends of the to-be-detected film are clamped and fixed through the fixed clamping assembly and the movable clamping assembly, the probability that a tearing opening is expanded to a non-detection part of the film after the film is tensioned and torn is reduced, and the output of plastic film waste during detection is reduced; in addition, through the cutting assembly and the secondary feeding mechanism, cutting of the unqualified part of the film and secondary feeding of the part to be detected are achieved.

The invention discloses a multi-agent automatic image retouching system based on content analysis, and the method comprises the steps: 1, receiving the input of an original image, calling a finely-adjusted multi-mode large model to carry out the combined semantic-visual analysis of the image, extracting key semantic elements in the image, and carrying out the recognition of the key semantic elements; comprising but not limited to background coordination degree, illumination condition, figure hair style definition, weather quality and composition layout rationality content information. Based on these information, the system performs comprehensive image quality scoring on the image, which combines image style, aesthetic level, definition and subjective quality multi-dimensional evaluation criteria. Meanwhile, based on a multi-dimensional image quality scoring result and understanding of image semantic content, the system automatically generates a repair and optimization strategy set for specific defects so as to guide a subsequent image quality enhancement process. The invention relates to the field of multi-modal model and multi-agent cooperation, and can meet the increasing requirements of rapid optimization and high-quality propagation of image contents.

The invention discloses a device and a method for intelligently investigating types and quantity of fishes. The device comprises a self-adaptive sonar detection module, a three-dimensional image acquisition module, an image preprocessing module, a fish detection module, a main body segmentation module and an identification and classification module. The sonar module is integrated with a three-frequency-band transducer, can dynamically switch frequency according to fish school depth, and realizes target tracking counting and quantity estimation in combination with an algorithm; the image acquisition module constructs a fish school three-dimensional image model by combining a depth camera with a light attenuation compensation algorithm; the preprocessing module fuses acoustic and optical features and expands a data set; the detection and segmentation module is used for accurately positioning fishes and removing impurities; the identification and classification module realizes type identification based on transfer learning and supports incremental learning of new fingerlings. According to the method, through cooperation of multiple modules, the problems of insufficient detection precision, poor image quality and the like in traditional investigation are solved, efficient and accurate investigation of fish species and quantity is realized, and technical support is provided for fishery resource management and ecological protection.