1821 results about "Stain" patented technology

The invention discloses an unmanned car washer stain panorama identification system. The system operation process specifically comprises the following steps: acquiring panorama image data of a target car; preprocessing the panoramic image data to obtain a standardized panoramic image set; performing stain area identification on the standardized panoramic image set based on a deep learning model to generate an initial stain distribution diagram; performing stain type classification on the initial stain distribution diagram according to a stain feature database to generate a stain classification result set; generating a dynamic cleaning path instruction set based on the stain classification result set and a cleaning strategy library; real-time images in the cleaning process are collected in real time, real-time stain residue analysis is conducted, and finally a cleaning effect feedback report is generated. The method has the following advantages and effects that the system of multi-dimensional stain feature recognition, classification and dynamic decision can be fused, so that the core contradiction that the cleaning strategy is not matched with the stain features in the prior art is solved.

The invention discloses a visual target tracking method and system for an unmanned aerial vehicle, and relates to the technical field of image acquisition and processing.The method comprises the steps that firstly, a bridge area is scanned in a directional mode in multiple different sun incident angle periods (the interval is larger than or equal to 30 degrees), and stain interference is eliminated through natural light multi-angle changes; the geometric stability of the crack edge is quantified by constructing a gradient direction distribution matrix of the time dimension; secondly, dividing multiple groups of scanning data into at least six direction groups, calculating pixel proportion variances group by group, accumulating the pixel proportion variances into a total variance parameter, and effectively distinguishing cracks from surface stains by combining a direction consistency verification mechanism (the total variance lt is judged as cracks by a preset threshold value); and finally, establishing a dynamic feedback closed loop based on a historical variance parameter sample library, adaptively adjusting a threshold value through a moving average value deviation ratio formula, and solving the problem of false detection caused by material aging and environmental fluctuation.

The invention discloses an automobile cushion sponge quality detection method based on artificial intelligence, and particularly relates to the field of quality detection.The method comprises the steps that sponge full-surface data is obtained and preprocessed, then detection units are divided according to double-layer grids, surface damage, stains and density uniformity are detected based on an artificial intelligence model, and preliminary screening is executed; if not, stopping the detection; after the primary screening is qualified, collecting internal structure data of the sponge, extracting defect characteristics, and carrying out a cyclic indentation test to obtain physical performance parameters; single-mode quality functions of the surface, the internal structure and the physical performance are constructed respectively, after function data are preprocessed and subjected to environment correction, primary filtering is performed firstly, then a D-S evidence theory is applied to perform multi-mode fusion decision so as to output a final quality judgment result, and meanwhile, detection data feedback optimization is combined to form a quality control closed loop, so that the quality control accuracy is improved. And intelligent and accurate detection of the quality of the automobile cushion sponge is realized.

The invention relates to the technical field of swimming pool cleaning equipment control, and provides a cleaning mode switching method and device for a multifunctional swimming pool cleaning robot. According to the method, water quality data corresponding to a swimming pool, stain distribution images of pool walls and pool bottoms, swimming pool structure size data and stain type data are collected in real time; dividing a cleaning area according to the swimming pool structure size data, generating a stain area marking result in the cleaning area in combination with the stain distribution image and the stain type data, and judging the pollution level of the whole swimming pool according to the water quality data so as to retrieve a corresponding cleaning mode from a preset cleaning mode library, the preset cleaning mode library comprises a global coverage mode, a key enhancement mode, a corner fitting mode and a water quality maintenance mode; and according to the called cleaning mode, a corresponding cleaning path plan and execution instruction is generated, and the driving module, the scrubbing module and the purification module are controlled to work cooperatively. According to the method, accurate analysis of the swimming pool environment is achieved, and blind cleaning of a traditional fixed mode is avoided.

The invention relates to the technical field of fin cleaning, in particular to a robot energy-saving system and method for intelligent control over the cold end of an indirect cooling auxiliary unit, and the system comprises the steps that a fin array surface temperature abnormal area and a stain coverage area are extracted, and a self-adaptive grid is divided according to the stain density; defining a local cleaning strategy and a global cleaning strategy based on the adaptive grid; generating a spiral traversal path based on an adaptive grid and a global cleaning strategy, and optimizing the shortest advancing path of the robot; and the robot moves in the working space according to the shortest advancing path, the local cleaning strategy is executed in sequence, and intelligent energy-saving cleaning of the fins of the indirect cooling auxiliary unit is achieved.

The invention relates to an unmanned aerial vehicle lens stain identification and cleaning method and system based on deep learning. The method comprises the following steps: generating an image acquisition instruction according to a set sampling period to obtain a to-be-detected image; the image acquisition instruction is used for instructing an unmanned aerial vehicle camera to perform image acquisition operation and feeding back in the form of a to-be-detected image; preprocessing the to-be-detected image to obtain a to-be-detected standard image; inputting a to-be-detected standard image into the stain detection model to obtain a stain identification result; the stain identification result comprises a central coordinate of the polluted area, bounding box data, a confidence coefficient score and a stain type label; generating a cleaning parameter and a cleaning instruction according to the stain identification result; the cleaning instruction is used for instructing the cleaning equipment to clean the unmanned aerial vehicle lens according to the cleaning parameters. By adopting the method, the problems of difficult stain identification and extensive cleaning decision in an existing unmanned aerial vehicle visual system can be solved, and the accuracy of stain detection and the intelligence of cleaning control are realized.

The invention provides a photovoltaic power station cleaning robot path planning method, medium and system, and belongs to the technical field of photovoltaic power station operation and maintaining.The photovoltaic power station cleaning robot path planning method includes the steps that firstly, a photovoltaic panel grid division model is established, and stain distribution information is obtained through the image processing technology to construct a stain distribution matrix; secondly, calculating a cleaning water demand matrix based on stain distribution, and meanwhile, considering an influence relationship among vertical column grid units; determining the cleaning duration by combining the efficiency of the cleaning device; constructing a robot energy consumption model; calculating an optimal cleaning path by applying a resource optimization distribution function; integrating the environmental parameters and inputting the environmental parameters into a photovoltaic cleaning neural network model to predict the actual cleaning difficulty; a cleaning strategy is dynamically adjusted accordingly; and executing the final path and recording data to update the related model. According to the method, through real-time sensing and analysis of the stain distribution characteristics, intelligence of path planning of the cleaning robot is realized, and the technical problem that the cleaning robot cannot perform intelligent path planning according to the surface stain distribution characteristics of the photovoltaic panel is solved.

The invention provides an environment-considered photovoltaic panel cleaning scheme determination method, and belongs to the technical field of photovoltaic panel cleaning.The method comprises the steps that firstly, real-time weather data and prediction data are obtained, and an environment stain correlation model is established; a multispectral camera is adopted to identify the types of stains on the surface of the photovoltaic panel and calculate a distribution map; calculating a cleaning index by combining the generating capacity data and the solar irradiation intensity, and determining a cleaning priority; a weather adaptability scoring system is constructed, and a robot operation environment threshold value is set; dividing power generation time periods according to the sunlight intensity and calculating an optimal cleaning time window; cleaning schemes are matched according to different stain types, and parameters are adjusted according to environmental conditions; calling a stain cleaning efficiency optimization function to carry out multi-objective optimization; and finally, generating an optimal cleaning path and parameter configuration by utilizing the photovoltaic cleaning planning network model. The technical problem that a photovoltaic panel cleaning scheme cannot be adaptively optimized according to complex and changeable environmental factors is solved.

The utility model discloses a cleaning needle for removing stains on the inner wall of a flared pipe. The cleaning needle is characterized by comprising a needle seat, a needle rod integrally connected with the needle seat, and a needle cone positioned at the end part of the needle rod, the appearance of the needle base is larger than the outer diameter of the needle rod, a compressed air injection hole is formed in the needle base and the needle rod in the axial direction, an air injection hole is formed in the bottom end of the compressed air injection hole, and the needle rod is longer than a flaring forming inner hole of a machining pipe. A series of problems of complicated production process, sewage and VOC discharge, low production efficiency, high cost and the like of traditional technologies such as water-based cleaning agent cleaning and solvent cleaning are solved, stains are cleaned, the structure is simple, one cleaning needle can be used for a long time, hardware such as a compressor can be integrated into other compressed air nets of a workshop after being put into one time, and the cost is low. No special operation environment requirement exists, and the comprehensive cost is low.

The invention discloses a high-speed strip steel defect detection method and system based on anomaly enhancement and image reconstruction, the problem that part of defects are difficult to recognize is solved by adopting a pseudo anomaly enhancement method, the defects difficult to detect are divided into a scratch type and a stain type, a local average gray level floating method is designed to generate scratch type pseudo anomalies, and the defect detection accuracy is improved. And a contour random disturbance method is used to generate stain type pseudo anomalies, and local difference features are provided for the model. And a lightweight image reconstruction model of an encoder-decoder structure is designed to achieve high detection speed and meet the requirement of synchronous detection. According to the method, performance testing is carried out on a strip steel surface image data set, advanced performance evaluation is obtained, and the effectiveness of the method is proved.

The invention discloses a stain identification method and equipment applied to a photovoltaic panel and a medium, belongs to the technical field of machine vision, and is used for solving the technical problem that stains on the photovoltaic panel cannot be accurately identified in the prior art. The method comprises the following steps: preprocessing an original image in a photovoltaic panel video to generate a preprocessed image; using a preset generative adversarial network model to generate a stain-free reference image according to the preprocessed image; extracting a part with the gray level higher than a preset gray level threshold value in the blot-free reference image as a mask, and reducing the gray level of a part corresponding to the mask in the preprocessed image to be below the preset gray level threshold value to obtain a secondary processing image; in the secondary processing image, extracting a highlight area of which the gray level is higher than a preset gray level threshold value, and segmenting the highlight area into a plurality of areas to be marked by using a preset area segmentation algorithm; and calculating a velocity vector corresponding to each to-be-marked area based on a sparse optical flow algorithm, and marking. Through the above steps, the method can achieve the precise recognition of the stains on the photovoltaic panel.

The invention relates to the technical field of unmanned aerial vehicle intelligent maintenance and cleaning, in particular to an unmanned aerial vehicle-based main road camera non-closure autonomous cleaning system and method, and the system executes the following steps: an unmanned aerial vehicle hovers at the outer side of a camera, collects millimeter wave-polarization-acoustic signals to generate stain spectrum fingerprints, and obtains cleaning parameters through optimization; and the model prediction controller sequentially executes plasma, acoustic cavitation and ion air curtain dry cleaning under energy constraint, self-adjusting parameter iteration is carried out when residues exceed a threshold value, and return flight and battery replacement are carried out after the standard is reached. And the cleaning progress, the residual index and the energy consumption are packaged and chained, the cloud self-learning model updates online and outputs a next order of instruction, and non-stop, low-consumption and efficient autonomous cleaning is realized.

The invention relates to the technical field of image data processing, in particular to a stain detection method for fresh egg cleaning, and the method comprises the steps: determining the gradient direction entropy of a pixel point according to the entropy sum of the number of pixel points located in each gradient direction angle interval in a pixel point initial window in a fresh egg gray image; according to the horizontal component sum and the vertical component sum of the gradient direction angles of all the pixel points in the pixel point initial window, calculating the local geometric direction consistency of the pixel points; calculating the size of a target window of the pixel point; the segmentation threshold value of the pixel point is calculated according to the gray average value and the gray standard deviation of all the pixel points in the pixel point target window, the fresh egg gray image is processed according to the segmentation threshold value of the pixel point, a fresh egg gray image stain detection result is obtained, and the accuracy of the fresh egg stain detection result is effectively improved.

The invention belongs to the technical field of oil stain detection, and particularly relates to a water surface oil stain detection method, system and equipment, the water surface oil stain detection system comprises a flexible circuit board FPC, an InGaAs infrared sensor, an infrared narrow-band filter and an optical lens, light enters a visible light camera and the optical lens through an external lens, the visible light camera collects visible light images, and the visible light images are transmitted to the InGaAs infrared sensor; the optical lens focuses a light beam, then light with a specific wavelength is screened out through the near-infrared narrow-band filter, the screened light is received by the InGaAs infrared sensor, and the InGaAs infrared sensor is connected with the FPC; the method has the advantages that visible light imaging and short wave infrared spectrum analysis technologies are ingeniously fused, so that the accuracy and the reliability of oil stain detection are remarkably improved. Under the current background that environmental protection and water resource management are increasingly severe, powerful technical support is undoubtedly provided for water surface pollution monitoring.

The invention provides a steel plate surface defect detection method and system based on multi-source data fusion, and relates to the technical field of steel production automatic detection.The method comprises the steps that water stain pretreatment is conducted on the surface of a steel plate, the residual humidity is reduced to be below a preset low humidity threshold value, and a dry surface is obtained; based on the dry surface, collecting multi-source data through a 3D laser line scanning camera and a 2D industrial line scanning camera which are synchronously triggered, obtaining line scanning laser point cloud data, a line scanning reflectivity gray level image, a line scanning depth gray level image and an area array reflectivity color image, and establishing a space corresponding relation among the multi-source data; and forming a multi-source data set based on the spatial correspondence among the multi-source data. According to the method, a high-precision and high-robustness steel plate surface defect detection system covering the whole detection process is constructed by eliminating water stain interference, unifying a multi-source data reference, optimizing data quality, accurately identifying defects and realizing automatic feedback and data closed loop.

The invention provides a textile surface defect detection method and system based on image features, and relates to the technical field of textile surface defect detection.The method comprises the steps that an RGB image of a to-be-detected textile is obtained and zoomed to 512 * 512 pixels, and then at least five semantic tags are output through a semantic segmentation model; the method comprises the following steps: adopting hyperspectral imaging with the wavelength of 400-1700 nm and the resolution of 10 nm, calculating the pixel reflectivity through a reflectivity formula, comparing the reflectivity of 5-8 characteristic wavebands, marking as a pollutant if the average deviation exceeds a threshold value, matching the type and position of a stain with a spectral angular distance, and after a standard sample and a semantic region are converted into an LAB color space, extracting a minimum rotation enclosing rectangle and dividing a grid, invalid pixels are filled with 3 * 3 neighborhood effective pixel weighted mean values, and a grid LAB mean value is calculated. A grid pollutant pixel corresponding to an image to be detected is filled with a standard sample in a mean value mode, then channels A and B are clustered through a sliding window, a pilling area is preliminarily judged according to the standard deviation of a channel L, and the pilling area is confirmed by combining the structured light three-dimensional point cloud and comparing with a threshold value.

The invention discloses an amyloid-like protein mediated base material surface controllable functional modification method, which comprises the following steps of: dip-coating or spraying an aqueous solution containing protein, a photoinitiator, an anti-fouling monomer and an adhesive monomer on the surface of a base material, and irradiating by ultraviolet rays to form a uniform anti-fouling amyloid protein nano coating on the surface of the base material. Functional monomers and amyloid protein self-assembly are combined, and the effects of adhesion and stubborn stain resistance can be achieved by simultaneously constructing a dual-adhesion unit and a functional anti-fouling unit. The protein nano coating has high optical permeability, on the premise that the original appearance of a base material is not affected, the functions of surface stubborn stain resistance, self-cleaning, stable adhesion and the like are added, stains on the surface of the coating can be washed away through simple water flow washing, and the protein nano coating has wide application prospects in daily life.

The invention relates to the technical field of automatic control, and provides an intelligent detergent putting control method and system for automatic car washing equipment. The method comprises the following steps: acquiring a vehicle surface stain image through a high-precision camera, and identifying stain features to generate a multi-dimensional feature set; performing clustering division on the feature set to obtain a stain area feature set, and calling a car washing scheme parameter space; analyzing the region feature set based on the parameter space, and determining cleaning scheme parameters of each region; a cleaning agent putting part control logic is associated, and area putting control parameters are generated; the environment temperature and humidity are monitored in real time, and cleaning agent putting and feedback regulation and control optimization are carried out in combination with control parameters, so that the technical problems that existing automatic car washing equipment lacks pertinence during cleaning agent putting, the cleaning agent putting amount cannot be accurately controlled according to the actual stain condition of a car and environment changes, the car washing effect is poor, and cleaning agents are wasted are solved. The technical effects of improving the car washing quality, reducing the waste of the cleaning agent and achieving intelligent and accurate control are achieved.

The invention relates to the technical field of oil stain detection, and discloses an oil-filled equipment oil stain detection method based on fluorescence reflection. The method comprises the following steps: carrying out cleaning pretreatment on a to-be-detected surface; an ultraviolet light source is utilized to excite the oil stain to generate fluorescence; collecting a fluorescence image through an imaging device with a filter; performing noise reduction and edge processing on the image to extract an oil stain area; and finally judging the existence and type of oil stain and estimating the thickness according to image features. According to the invention, the fluorescence characteristic of the oil stain is utilized, no fluorescent agent needs to be added, and pollution-free, high-precision and high-efficiency non-contact detection is realized.

The invention discloses an electronic cleaning equipment operation optimization method and system based on the Internet of Things, and belongs to the field of control systems.The electronic cleaning equipment operation optimization method based on the Internet of Things comprises the following steps that ground stains are sensed in real time through a multi-mode sensor and an edge computing technology, and common stains and excrement stains are distinguished; after excrement stains are recognized, the electronic cleaning equipment is controlled to stop sweeping immediately, a rolling brush / dust collection module is closed, and stain diffusion is avoided; through cooperation of the Internet of Things and family environment data, a risk area is pre-judged, and cleaning logic is adjusted; compared with the prior art, the method has the beneficial effects that the dirt is classified, and when the dirt is judged to be excrement dirt, the electronic cleaning equipment is controlled to stop cleaning, so that the excrement dirt is prevented from being diffused; meanwhile, through cooperation of the Internet of Things and family environment data, the risk area is pre-judged, the cleaning logic is adjusted, the cleaning frequency and path of the electronic cleaning equipment are dynamically changed, and the stain diffusion risk is reduced.

The invention discloses a power equipment infrared detection method based on an image, and relates to the technical field of image detection, and the method comprises the steps: obtaining an infrared thermal image, a visible light image and a vibration data set of to-be-detected equipment; establishing a displacement prediction model to obtain an alignment degree and a first stain image; calculating vibration energy density according to the vibration data set, and calculating thermal gradient variance according to the infrared thermal image; according to the first stain image morphological analysis and the vibration energy density, obtaining an adhesion index; determining a main boundary and a diffusion boundary of the stain region based on the adhesion index and the thermal gradient variance, evaluating an influence value of the cross region, and determining a generation strategy based on the influence value to obtain a second stain image; vibration data are collected in real time, dynamic changes of equipment and stains in a vibration mode are effectively recognized, and image fusion is dynamically regulated and controlled according to the changes; and a plurality of alignment indexes are defined, so that the accuracy of stain image recognition is further improved, and the precision of equipment monitoring is improved.

The invention provides a control method and system of a sweeping robot, and the method comprises the steps: obtaining image data, three-dimensional space data and sound data of a surrounding environment, and extracting object attribute features, space features and sound attribute features in the environment; fusing the object attribute feature, the spatial feature and the sound attribute feature to obtain an identification result; based on an identification result, executing hierarchical scene adaptive planning and executing a cleaning task by adopting a self-adaptive variational cleaning strategy; after the cleaning task is completed, second image data of the cleaned area are collected; recognizing residual stains on the ground according to the second image data and a preset cleaning effect recognition model; and if the area of the residual stain exceeds a preset area threshold value, triggering a re-cleaning program, and automatically switching a corresponding cleaning mode according to the type of the residual stain. According to the scheme, when the sweeping robot encounters the obstacles in the sweeping area, the sweeping robot can intelligently take accurate countermeasures, and an excellent sweeping effect is achieved.

Techniques for computationally generating a faux-stain image of an unstained sample using intensity measurements of visible light and ultraviolet light through, or reflected by, the unstained sample. In some cases, the faux-stain image is provided as input to a trained deep neural network and an outcome prediction such as likelihood of metastasis is determined based on output of the trained deep neural network.

The invention discloses a surface treatment process of a hydraulic rod and a coating monitoring system, relates to the technical field of electroplating, and aims to solve the problems that various complementary sensing technologies cannot be integrated, information such as thickness, uniformity and microstructure cannot be synchronously acquired in a lossless, in-situ and real-time manner, and closed-loop control cannot be formed based on the information in the background technology. According to the scheme, the method comprises the steps that PR electrolytic degreasing is conducted, a workpiece serves as an anode, an oil film is torn through the oxygen evolution effect, an extremely-thin oxidation film possibly existing on the surface of the workpiece is dissolved, base metal is completely activated, and an optimal combination basis is provided for a follow-up coating; and washing with hot water, namely removing residual alkaline degreasing fluid and emulsified oil stains by using hot water with the temperature higher than 60 DEG C in a stirring and spraying manner. According to the method, multi-dimensional information such as the thickness, the uniformity and the microstructure can be synchronously obtained in a lossless, in-situ and real-time mode in the surface treatment process, closed-loop control is formed based on the information, and it is ensured that plating growth is always located on the optimal track.

The invention relates to the technical field of unmanned aerial vehicle cleaning, and discloses a cleaning application and data processing system suitable for an unmanned aerial vehicle, and the system comprises a stain distribution modeling module, a sensor configuration module, a task definition module, an execution control module, an analysis feedback module and the like. According to the system, a stain distribution model is constructed through multispectral imaging, detection points are arranged to calculate the stain density gradient, cleaning indexes are defined according to physical parameters, steady-state and dynamic response and composite cleaning are executed, and abnormal tracing and strategy updating are conducted on areas which do not pass through cleaning. In addition, the system also comprises a data aggregation and communication relay module. The system can accurately model, scientifically regulate and control cleaning parameters, adapt to complex working conditions, improve the cleaning efficiency and quality and is suitable for the field of unmanned aerial vehicle cleaning.

The invention provides an engineering machinery man-machine binding card punching management system based on fingerprint biological recognition, and belongs to the field of fingerprint recognition, and the method comprises the steps: obtaining a fingerprint image of a to-be-recognized constructor and an initial fingerprint matching degree; dividing the fingerprint image of the construction personnel to be identified into a plurality of areas; obtaining a normal fingerprint area and a suspected stain area according to the shape of each area; according to the number of normal fingerprint areas around each suspected dirty area and the distribution of pixel points in each suspected dirty area, obtaining a dirty area; and completing identification according to the normal fingerprint area, the stain area and the initial fingerprint matching degree of the to-be-identified constructor. The invention aims to solve the problem that the construction personnel cannot identify the fingerprint when performing fingerprint card punching identification on the construction site due to the fact that the fingers of the constructor may be stained with oil stains, soil and the like in the construction process. And the problem that fingerprint card punching comparison is wrong may occur.

The invention discloses an electrolytic capacitor pin defect visual inspection device, and relates to the technical field of visual inspection devices. Through cooperation of the first electric telescopic rod, the material guide plate and the triangular plate, the conveying angle and position of the capacitor can be adjusted conveniently, and the flexibility of capacitor conveying path adjustment is improved; the analysis module carries out continuity judgment and length comparison on the gray blocks, the length edge of the capacitor in the image can be accurately recognized, even if part of stains exist or the gray blocks are not continuous due to uneven illumination, the complete length edge can be obtained through extension processing, edge recognition errors caused by local defects of the image are effectively avoided, and the image quality is improved. The position judgment precision of the capacitor is improved; by dynamically calculating the deviation angle and correspondingly adjusting the exposure time and gain, the detection device can automatically adapt to different skew states of the capacitor, manual intervention is not needed, and the automation degree of detection and the universality of the device are improved.

The invention discloses a sweeping robot cleaning control system with a self-adaptive mode adjustment function, and relates to the technical field of intelligent cleaning robot control systems, and the system comprises a multi-mode sensing module, a central processing module and a cleaning execution module. The multi-modal sensing module collects multi-dimensional data such as a hyperspectral image, acoustic response and gas concentration of ground stains, and the central processing module extracts features through the data fusion unit and deeply fuses the features to generate analysis results of stain types, adhesion strength and chemical components. The decision control unit dynamically generates a cleaning instruction in combination with a preset strategy and the real-time state of the robot, the cleaning execution module executes a targeted cleaning action, and meanwhile, the system constructs a dynamic stain atlas to achieve historical data reuse and prospective adjustment, so that the cleaning efficiency and the resource utilization efficiency are remarkably improved, and the cleaning effect is improved. The method is suitable for intelligent cleaning scenes of home and office environments.

A two-in-one solid detergent composition includes the following raw material components based on 100% weight: 10% to 30% of (a) a forming agent, 10% to 50% of (b) a surfactant, 5% to 20% of (c) a softener, 10% to 30% of (d) a forming aid, and optionally 0.1% to 1% of (e) a softening conditioner, 0.1% to 1% of (f) a forming stabilizer, 1% to 10% of (g) a water softener, 1% to 10% of (h) a stain solubilizer or 0.1% to 5% of (i) a flavoring agent, and balance of deionized water.

The invention provides a strip steel defect detection method and device and electronic equipment, and the method comprises the steps: firstly detecting a defect region in a strip steel surface image, cutting out a defect region block, recording the position coordinate of the defect region block, setting a defect classification model, dividing a roller mark defect and a stain defect into the same target category, and carrying out the detection of the defect. And a defect classification model is adopted to determine a defect category corresponding to each defect area block, then position periodical detection is performed on the defect area blocks in the target category, the defect areas conforming to the position periodicity are determined as roll mark defects, and the defect areas not conforming to the position periodicity are determined as stain defects. According to the scheme, stain defects, roll mark defects and other various types of defects can be accurately distinguished in an image processing mode under the condition that stains on the surface of the strip steel are not wiped, so that the defects caused by the fact that the stains on the surface of the strip steel need to be removed firstly and then the defect types need to be detected in the prior art are overcome; and the detection of the roller mark defects is not interfered by other defect types, so that the detection accuracy is higher.