161results about How to "Efficient and accurate detection" patented technology

The invention provides a deep counterfeit detection method based on self-mixing, and compared with the prior art, the method generates a mixed image containing counterfeit artifacts by using a self-mixing technology, and carries out the precise detection and recognition of an artifact region of an input image through combining a backbone network OfficientNet-b3 and a counterfeit artifact detection module. The method comprises the following steps of: firstly, performing framing processing on a real image, generating a source image and a target image by utilizing a self-mixing technology, and generating a mixed image with counterfeit artifacts by combining a mask generator with an alpha blowing model; and then, the backbone network extracts global features from the mixed image, and the fake artifact detection module reinforces local feature expression of a mixed region through an artifact space cooperation attention module, so that an artifact region is accurately positioned. And finally, fusing global and local features by using residual connection, and transmitting the fused features to a full connection layer to complete classification. According to the method, various forged image types can be effectively detected, and excellent generalization ability is shown in cross-dataset and unknown scenes.

The invention relates to the technical field of composite material surface defect detection, in particular to a composite material surface defect detection system and method.The method comprises the steps that collected composite material surface images are input into a convolutional neural network model, a support vector machine model and a random forest model; each model independently carries out defect detection on an input composite material surface image, outputs a probability group containing the probability of each defect category, generates an evaluation label for the probability group, ensures that the evaluation module can detect whether data loss occurs in the transmission process, then obtains the defect category probability output by each model, and determines whether the data loss occurs or not. And respectively carrying out weighted fusion on the probability of each category by using a weighted average formula, and determining the category of the surface defect of the composite material based on a set threshold value. The capturing capability of the system on different types of defect characteristics in composite material surface defect detection is guaranteed, and the beneficial technical effect of accurately and efficiently detecting the composite material surface defects is achieved.

The invention discloses a device and method for detecting technological parameters of a part, and relates to the field related to automatic control, and the device comprises a part information obtaining module which is used for carrying out image collection and obtaining part laser point cloud information; the part space model acquisition module is used for performing space registration fusion on the part image information and the part laser point cloud information to obtain a standard part space model; the multi-dimensional loss analysis module is used for performing multi-dimensional loss analysis on the target and standard part space models; the part process detection characteristic parameter acquisition module is used for acquiring internal defect characteristic parameters and part process detection characteristic parameters; and the part production optimization control module is used for carrying out process parameter compensation optimization and carrying out production control through part production process optimization parameters. The technical problems that existing part technological parameter detection is insufficient in detection precision and low in detection efficiency are solved, and the technical effects that the parts are efficiently and accurately detected, and adaptive optimization of the production technology is achieved are achieved.

The invention relates to the technical field of railway line flaw detection, in particular to railway line automatic comprehensive flaw detection equipment which comprises two steel rails and a flaw detection car and further comprises a C-shaped mounting plate, a buffer assembly, a waist detection mechanism and a bottom detection mechanism. A first fixing rod is transversely arranged in the C-shaped mounting plate, fixing sleeves are slidably connected to the two ends of the first fixing rod correspondingly, and balance assemblies are arranged at the ends, away from each other, of the two fixing sleeves correspondingly. The device has the advantages that different detection assemblies are arranged for different detection areas, so that steel rail detection of different parts is more efficient and accurate, it is ensured that different types of defects can be detected, the detection period is shortened, the balance assembly and the buffer assembly cooperate with each other to enable the head detection assembly to keep a dynamic balance state, and the detection efficiency is improved. The probe is effectively protected from vibration damage, and the adaptability of the probe to the change of the steel rail and the detection stability of the probe are improved.

The invention discloses a multi-dimensional rumor detection and prevention system based on a large language model, which relates to the technical field of rumor detection and comprises a data collection module, a data preprocessing module, a feature extraction module, a background knowledge integration module, a logical reasoning module, a detection and classification module and a prevention and response module. According to the rumor detection method and system, all aspects of the rumor can be comprehensively considered from multiple angles, the aspects include but are not limited to the content of the rumor, the propagation path of the rumor and the interaction condition between the users, and through the comprehensive analysis, the system can ensure that detection of the rumor covers all related dimensions, so that the accuracy and comprehensiveness of detection are improved, and the detection efficiency is improved. The rumor detection and classification work can be efficiently carried out by utilizing an advanced artificial intelligence technology, and through an intelligent means, the system not only improves the detection efficiency, but also remarkably improves the detection accuracy and the intelligent level, so that the rumor detection is more accurate and efficient.

The invention provides a method for testing the skin migration volume of 20 organophosphorus flame retardants in cosmetic sponge and application, and the method comprises the following steps: fixing the cosmetic sponge containing the organophosphorus flame retardants on an electric powder puff device, and beating an artificial film at a set beating frequency by using the electric powder puff device; cutting the artificial membrane into pieces, carrying out ultrasonic extraction by adopting acetonitrile, and filtering the extract to obtain a to-be-detected solution; placing the to-be-detected liquid in liquid chromatography-tandem mass spectrometry to detect the test content of the organophosphorus flame retardant in the artificial film; the skin migration amount of the organophosphorus flame retardant is calculated based on the test content of the organophosphorus flame retardant in the artificial film, the volume of the to-be-tested liquid and the area of the artificial film, the problem that the skin migration amount is lost in skin exposure risk evaluation of the products is solved, and accurate evaluation of the skin exposure risk of harmful substances in cosmetic sponge products is facilitated.

The invention relates to a multi-channel flying patting inspection machine and a circuit board inspection method. The inspection machine comprises an equipment frame, wherein a feeding station, a detection station, a disc stacking station and a labeling station are sequentially arranged on the equipment frame; the multiple feeding devices are arranged at the feeding station side by side; the plurality of flying shooting detection devices are arranged at the detection station side by side and are arranged in one-to-one correspondence with the plurality of feeding devices; the plurality of tray caching and stacking devices are arranged at the tray stacking station side by side and are arranged in one-to-one correspondence with the plurality of fly-shot detection devices; and the labeling device is arranged at the labeling station and is in butt joint with the plurality of tray caching and stacking devices. The problems that in the prior art, manual feeding, disc stacking and labeling are low in efficiency and high in cost, the risks of material mixing and missing detection exist, and the mode that multiple sets of cameras are adopted for shooting the whole disc is not suitable for detection of circuit board products with laser two-dimensional codes can be solved.

The utility model provides a tab detection device and a battery production line, the tab detection device is used for detecting tabs in an electrode assembly, and the tab detection device comprises a clamping mechanism, a visual module and a control device. The clamping mechanism comprises a support, a first clamping assembly and a second clamping assembly. The first clamping assembly used for clamping the electrode assembly is connected to the support and provides clamping force in the first direction. The second clamping assembly used for clamping the end cover is connected relative to the first clamping assembly in a reciprocating motion mode in the second direction, the second clamping assembly provides clamping force in the first direction, the first direction is perpendicular to the second direction, and the second direction is perpendicular to the end cover. The control device controls the second clamping assembly to clamp the end cover in the state that the first clamping assembly clamps the electrode assembly, the second clamping assembly moves in the direction where the end cover is far away from the electrode assembly in the second direction, the visual module is controlled to shoot the tab, and the control device conducts detection based on an image shot by the visual module. According to the invention, whether the bent tab has defects can be accurately and efficiently detected.

The invention belongs to the technical field of shelf structure detection, and particularly discloses a shelf structure deformation detection device and method based on laser-assisted calibration, and the device comprises a projection plane, a reference assembly, a mapping assembly, an image collection assembly, and a connection relation thereof. Acquiring a mapping light spot and a reference line image in real time through an optical camera of an image acquisition assembly, and acquiring light spot distribution information relative to a reference mark; and calculating the space attitude of the detection position of the goods shelf structure to realize the detection of the deformation of the goods shelf structure. According to the invention, the three-dimensional space characteristics of the shelf structure deformation are projected to the two-dimensional plane, the image information projected by the laser is acquired through the image acquisition assembly, the shelf structure deformation condition is analyzed, the shelf structure deformation can be efficiently and accurately detected on the premise of not influencing the original function design of the warehouse, and technical support is provided for safety monitoring of the warehouse shelf.

The invention relates to a deep learning-based Hamp; the invention discloses an E image ovarian cancer polyploid tumor giant cell (PGCCs) detection method. The method comprises the steps of specific data set construction, image annotation and data division, feature learning and optimization by using an OCDet model, and model training and evaluation. The method comprises the following steps: firstly, establishing a Hamp containing PGCCs; and E, dyeing the image data set, then accurately marking the image and dividing the image into training, verification and test sets. The OCDet model takes CSPDarkNet as a core, combines an ECA mechanism, and focuses on deep learning and recoding of pathological semantic features. By training a data set, the model performs parameter updating and optimization through back propagation and gradient descent to identify PGCCs features. The verification set is used for model optimization, and the test set is used for evaluating model performance. According to the automatic detection technology, doctors can be helped to improve the diagnosis efficiency and reduce errors, important reference is provided for clinical treatment and prognosis evaluation, and remarkable clinical application value is shown.

The invention discloses a road inspection method based on data fusion and edge calculation, and relates to the field of road inspection, and the method comprises the steps: obtaining a pre-marked weak signal road section according to a preset road inspection route; acquiring real-time point cloud picture information and real-time driving information before the inspection unmanned vehicle enters the weak signal road section, constructing a road simulation model and a driving state model, and updating the road simulation model and the driving state model in real time after the inspection unmanned vehicle enters the weak signal road section; carrying out model fusion on the updated road simulation model and the driving state model to obtain a road driving model; acquiring real-time environment image information after the inspection unmanned vehicle enters the weak signal road section; and determining disease feature types according to the real-time point cloud image information and the real-time environment image information, and obtaining road section disease position information of the weak signal road section based on the disease feature types and a road driving model. The method has the effect of efficiently and accurately detecting the types of the pavement diseases and the accurate positions of the pavement diseases.

The invention discloses a discharge source positioning method, device and system and a nonvolatile storage medium. The method comprises the following steps: acquiring a plurality of high-frequency pulse electric signals respectively representing discharge characteristics at a plurality of positions; based on the multiple high-frequency pulse electric signals, whether the discharge position of the discharge source is located in the transformer is judged; and under the condition that the discharge position is located in the transformer, ultrasonic signals are collected, and a three-dimensional space positioning strategy is adopted, and the multiple high-frequency pulse electric signals and the ultrasonic signals are combined to determine the space three-dimensional coordinates of the discharge source. According to the invention, the technical problem of low efficiency of transformer partial discharge detection and positioning is solved.

The invention belongs to the related technical field of surface defect detection, and discloses a large sheet metal part punching defect detection method and system based on deep learning, and the method comprises the steps: (1) collecting a multi-view image of a to-be-detected sheet metal part; (2) training a deep learning hole site target detection model by using the image, and further obtaining position data of a hole site on the to-be-detected sheet metal part based on the model and the acquired image; (3) hole site matching between different camera images is carried out through a multi-view geometric principle, and repeated hole sites in different images are detected; and (4) removing repeated hole sites from all the hole sites obtained by the deep learning hole site target detection model, counting, comparing an obtained counting result with the number of designed hole sites or comparing the positions of the hole sites with corresponding hole site design positions, and further judging whether hole site punching omission exists or not according to a comparison result. According to the invention, the real-time performance and the accuracy of the missed punching detection of the stamping part are improved.

The present invention provides a consistency detection method for a secondary battery, comprising: providing a battery to be tested and a plurality of reference batteries; obtaining the Nyquist spectrum of each reference battery; obtaining the first characteristic frequency in the Nyquist spectrum of each reference battery, the first characteristic frequency being the frequency corresponding to the ohmic internal resistance of the reference battery in the Nyquist spectrum; obtaining a first average characteristic frequency according to the first characteristic frequency of each reference battery; using a first test current of the first average characteristic frequency to input the battery to be tested, and obtaining a corresponding first response voltage; obtaining the ohmic internal resistance of the battery to be tested according to the first response voltage and the first test current of the first average characteristic frequency, and judging the consistency of the battery to be tested by the ohmic internal resistance of the battery to be tested. The consistency detection method provided by the present invention can efficiently and accurately detect the consistency of a secondary battery.

The invention relates to a large language model processing method and device, equipment, a storage medium and a program product, and relates to the technical field of natural language processing. According to the method and the device, the adjusted large language model can be used for expanding the risk data set, and the original generation capability of the large language model can be prevented from being degraded. The method comprises the following steps: identifying security activation parameters in a large language model for a sample by using data; inputting the multiple groups of risk data into the large language model and recording a current gradient slice corresponding to the security activation parameter after the large language model is processed; according to the similarity between the current gradient slice and the abnormal gradient slice, identifying valid data in the multiple groups of risk data to obtain a data training set; performing low-rank fine tuning on the large language model according to the data training set to obtain fine tuning parameters; and updating the original weight of the security activation parameter based on the fine tuning parameter and the security activation parameter, and adjusting the model parameter according to the updated weight, the adjusted large language model being used for expanding the risk data set.

The invention belongs to the technical field of livestock farms, and particularly relates to an intelligent livestock farm harmful gas monitoring device which comprises a base station with a display, and further comprises an unmanned aerial vehicle, a charging mechanism, a collection device, a detection device and a cleaning device which are mounted on the base station. According to the intelligent harmful gas monitoring device for the livestock farm, the detection device is arranged, collected air can be detected, after gas in a collection pipe is pushed by a pushing piston to extrude a sealing separation blade in a detection bottle to descend, the collected gas enters the detection bottle through an air inlet hole of the detection bottle, and the detection effect is improved; the gas is detected through the sensor group, and the detected result can be transmitted to the display screen, so that the gas in the livestock farm is monitored, the gas detection is more efficient and accurate, harmful gas in the air in the livestock farm can be effectively monitored, and the safety and health of the livestock farm environment are ensured; and the filtering assembly can filter harmful gas in the detected air.

The invention relates to the field of equipment detection, in particular to an oil leakage detection system and method for oil charging equipment. The system comprises a hydrocarbon sensing cable, the center of the hydrocarbon sensing cable is provided with a sensing line, the outer side of the sensing line is wrapped with an adsorption part, the adsorption part is used for adsorbing leaked liquid hydrocarbon, and the sensing line is used for detecting resistance change; the hydrocarbon sensor is connected with the hydrocarbon sensing cable and is used for converting resistance change caused by adsorbed hydrocarbon into a digital signal and sending the digital signal to the sensor module; the sensor module communicates with the background system monitoring host and is used for detecting the oil leakage condition of the oil charging equipment; the background system monitoring host is used for receiving and processing signals sent by the sensor module. And missed inspection caused by insufficient manual inspection is reduced. According to the invention, the plurality of sensor modules are used for covering key parts of the oil filling equipment, especially areas which are difficult to observe, so that dead-angle-free monitoring is ensured.

The invention relates to the technical field of image segmentation, and particularly discloses a bridge crack segmentation method based on YOLOv8n-seg, which comprises the following steps: acquiring a bridge crack image, labeling the acquired bridge crack image and making a data set, respectively adding a C2fDCNv2 module and a CBAM attention mechanism into a backbone network and a detection head of a YOLOv8n-seg model to obtain an improved YOLOv8n-seg model, and segmenting the bridge crack by using the improved YOLOv8n-seg model. And under the supervision of a loss function, training the improved YOLOv8n-seg model by using the data set to obtain a bridge crack segmentation model, re-collecting a bridge crack image, and inputting the bridge crack image into the bridge crack segmentation model to obtain a segmentation result which is an image marking all bridge cracks. According to the invention, the detection precision of small targets such as bridge cracks is high, the manual labor is reduced, the cost is low, and the bridge cracks can be efficiently and accurately detected.

The utility model provides battery riveting detection equipment, and the equipment comprises a bearing mechanism in which a to-be-detected battery is disposed; the fixing mechanism comprises a pressing driver and a rotary pressing head; the detection mechanism comprises a first detector and a second detector, and the first detector and the second detector are symmetrically arranged in the thickness direction of the battery; and the control assembly comprises a data processing module, and the first detector and the second detector are connected with the data processing module. In the actual operation process, operation is easy and convenient, detection is accurate, the production cost is low, most importantly, the device can be matched with a riveting production line at the present stage, seamless connection of riveting machining and quality detection is achieved, therefore, the detection efficiency is greatly improved, the labor cost is reduced, and the product quality is improved. Compared with a common battery detection technology at the present stage, the method has the remarkable advantages of being flexible and convenient to use, efficient and accurate in detection, low in cost, wide in use scene and the like.

The invention discloses a high-efficiency solid state disk master control chip, which relates to the technical field of solid state disk master control chips, and comprises a central processing unit and a flash memory controller, the flash memory controller comprises a communication module, a flash memory control circuit and a comparison circuit; in the invention, in performance, data bus and instruction processing are optimized, read-write is accelerated, delay is reduced, and high-speed storage requirements are met; in the aspect of data, the flash memory control circuit accurately reads and writes, the comparison circuit corrects errors in real time, data reliability is guaranteed, and the problem of flash memory management is avoided; during bad block detection, factory leaving and newly-added bad blocks are accurately judged by means of multiple means such as specific verification data after all pieces are erased, and the central processing unit intelligently judges, saves time and improves efficiency; in hardware collaboration, a data selector intelligently switches paths, static random access and a memory guarantee data stability, a main state machine accurately controls a time sequence, stability is enhanced, and a high-quality storage scheme is provided for the solid state disk.

The invention provides a lightweight small target pest detection method and device and electronic equipment, and the method comprises the steps: carrying out the preprocessing of collected small target pest images, so as to construct a pest image data set; based on a YOLOv5 model, constructing a small target pest detection model; wherein a small target feature enhancement module is added in a trunk part of the YOLOv5s model, and the small target feature enhancement module is constructed based on a MobileNetv3 structure and a BiFomer structure; a VoV-GSCSP module is added to the neck part of the YOLOv5s model; training the small target pest detection model to obtain a trained small target pest detection model; and inputting the pest image data set into the trained small target pest detection model to obtain a detection result of the small target pest image. Through adoption of the method, the problems that the model complexity is high, the parameter quantity is large and the deployment of the model on edge equipment is not facilitated due to pursuing of relatively high recognition precision in a pest image detection method in the related technology are solved.

The present invention discloses a strip defect detection method based on deep learning. The encoder is based on U‑Net. The encoder extracts features in the image through repeated convolution operations, and performs five downsampling operations to extract features of different scales. The decoder part is symmetrical with the encoder. It receives the output of the encoder and performs five upsampling operations, and the result obtained by each upsampling will be spliced ​​with the feature map of the same scale of the encoder. A Gaussian low-pass filter is used before downsampling to remove high-frequency information therein to prevent aliasing due to the high original frequency of the feature map, which leads to deviation in the feature map. The removed high-frequency information contains detailed information of the original image, which is crucial to the detection of small defects in the strip, and directly discarding it will affect the effect. This method further adds these high-frequency information back to the feature map of the model decoder to produce a more refined segmentation effect.

The invention discloses a vehicle hub threaded hole size automatic detection device, and relates to the technical field of detection equipment, the vehicle hub threaded hole size automatic detection device comprises a base and a hub member, the top of the base is rotatably connected with a rotating disc, the axis of one side of the rotating disc is provided with a detection assembly, and the outer side of the detection assembly is provided with a clamping block for clamping and fixing the hub member. According to the automatic detection equipment for the size of the threaded hole of the vehicle hub, by introducing an alarm lamp and a circuit closing mechanism, when the size of the threaded hole of the hub does not meet a preset standard, a movable rod moves to an inclined area, so that a connecting copper sheet is in contact with a fixed copper sheet, a closed circuit is formed, and the alarm lamp is activated; the real-time alarm function can find and process unqualified hub parts in time, and the problem in the subsequent assembly or use process caused by the unqualified size of the threaded hole is avoided.

The invention provides a KASP molecular marker related to the ear length character of corn under the drought condition and application of the KASP molecular marker, and belongs to the technical field of molecular genetics. An SNP marker site which is related to the ear length character under the drought condition and has polymorphism is obtained through genome-wide association analysis; the SNP marker site references that a corn B73 RefGenv4 genome has nucleotide polymorphism C / T at a position 28374318 of a chromosome 2, when T is a dominant character, and nucleotide sequences for amplifying a KASP molecular marker primer group are as shown in SEQ ID NO.1-3. The novel molecular marker for the corn ear length is developed, genetic typing can be rapidly carried out by utilizing a KASP typing technology, detection is accurate and efficient, amplification is convenient and stable, and the molecular marker can be used for molecular marker-assisted selection and plays a very important role in predicting, identifying or assisting in identifying the ear length or breeding a new corn variety with the relatively long ear length.

The utility model relates to the technical field of image acquisition, in particular to a chip glass substrate image data acquisition device. Comprising a longitudinal movement guide rail, a sample bearing base plate is arranged above the longitudinal movement guide rail in a sliding mode, a transverse movement guide rail is arranged above the sample bearing base plate, and a first detection moving rail and a second detection moving rail are fixed to the front side of the transverse movement guide rail; a first sensor is fixed to the front side of the first detection moving rail, and a second sensor, a third sensor and a fourth sensor are fixed to the front side of the second detection moving rail. According to the device, the chip glass substrate can be automatically detected, and the situation that the through hole of the glass substrate is abnormal and blocked can be identified.

The embodiment of the invention provides a defect detection method and device, computer equipment and a storage medium, and belongs to the technical field of defect detection. The method comprises the following steps: acquiring training sample data, and extracting a sample defect image based on a defect region extraction model to obtain a sample defect positioning image; detecting the sample defect positioning image based on the image defect detection model to obtain a sample detection result; calculating a first loss value based on the sample defect mask data and the predicted defect mask data, and calculating a second loss value based on the sample defect category, the predicted defect category, the sample bounding box, and the predicted bounding box to determine a model loss value; updating model parameters of the defect area extraction model and the image defect detection model based on the model loss value so as to construct a target defect detection model; and detecting the target detection image based on the target defect detection model to obtain a target detection result. According to the embodiment of the invention, the defect detection efficiency and accuracy of the to-be-detected object can be improved.

The application discloses an aggregation-induced emission metal organic framework, a preparation method thereof and fluorescent / colorimetric dual-mode intelligent detection for trace HCl. 2+ The aggregation-induced emission metal organic framework is prepared by providing Zn(NO3)3.6H2O as Zn 2+ , imidazole-2-formaldehyde and an AIE molecule HDBB as organic ligands, and through a solvothermal method to obtain ZIF-90-HDBB, which can change the aggregation state fluorescence from yellow green to orange red after being contacted with HCl gas; the ZIF-90-HDBB is uniformly loaded on cellulose filter paper to obtain a paper-based sensor for portable detection of HCl, and through combination with a color recognition APP of a smart phone, the HCl can be rapidly and sensitively detected on site in a fluorescent / colorimetric dual mode. Since the HDBB molecule is prone to protonation under acidic conditions, the fluorescence color changes, and therefore the material can be accurately and efficiently used for detection of acidic gas, and the material can be recycled.

The utility model discloses a device for detecting parallelism and symmetry of a riveting pin workpiece, which relates to the technical field of riveting pin workpiece detection and comprises a detection workbench, a connecting rod fixing component is arranged at the upper end of the detection workbench, and a parallelism and symmetry detection component is fixedly clamped at the inner end of the connecting rod fixing component by the connecting rod fixing component; the end, away from the connecting rod fixing component, of the parallelism and symmetry degree detection component is inserted into an embedding groove in the top end of the pin riveting workpiece, a positioning cavity groove is formed in the other side of the upper end of the detection workbench, and a square clamping block at the bottom end of the pin riveting workpiece is sleeved with the positioning cavity groove. The parallelism and symmetry degree of the riveting pin workpiece are detected through the parallelism and symmetry degree detection part, and the parallelism and symmetry degree detection part is smoothly inserted into the caulking groove of the riveting pin workpiece, so that the caulking groove of the riveting pin workpiece is parallel to the square clamping block; two sides of the parallelism and symmetry degree detection component are symmetrical after being inserted into the caulking groove, which indicates that the caulking groove of the riveting pin workpiece has symmetry.

The invention relates to an auditory attention detection method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring an electroencephalogram signal without voice stimulation; converting the electroencephalogram signals into weighted pulse electroencephalogram features in an attention model; performing feature extraction on a sequence formed by the weighted pulse electroencephalogram features through a long-short term memory network model to obtain an electroencephalogram hiding state; classifying according to the electroencephalogram hiding state to obtain auditory attention information; the auditory attention information represents the sound source position of the voice felt by the object to which the electroencephalogram signal without voice stimulation belongs. By adopting the method, the calculation cost can be reduced on the basis of ensuring the accuracy.

The invention provides a vehicle detection method and a related device. The vehicle detection method comprises the following steps: acquiring left front detection data and right front detection data of a target vehicle; determining an abnormal detection item in the left front area according to the left front detection data; determining an abnormal detection item in the right front area according to the right front detection data; if the number of the abnormal detection items in the left front area meets a first water soaking number threshold value and the number of the abnormal detection items in the right front area meets a second water soaking number threshold value, it is indicated that water inflow defects exist in the left front area and the right front area of the target vehicle, and the target vehicle is determined to be a water soaking vehicle. According to the method, the left front area and the right front area of the cockpit of the target vehicle are detected respectively, and compared with detection of all areas of the cockpit of the target vehicle, whether the vehicle is the water soaking vehicle or not can be detected efficiently and accurately.