141 results about "Pyramid (image processing)" patented technology

The invention belongs to the technical field of computer vision and image processing, and discloses a remote sensing small sample target detection method based on double-attention guided transfer learning, and the method comprises the steps: obtaining a remote sensing image data set, and carrying out the preprocessing; taking the preprocessed remote sensing image training set as input, constructing a basic detection model by using a ResNet-101 backbone network, a feature pyramid network and a content awareness upsampling and regional proposal network, and obtaining basic model parameters; basic model parameters are used as input, a DA-FSDET network is trained based on a content awareness strip pyramid and a deformable attention area proposal network, and the trained DA-FSDET network is used to acquire a category detection frame containing small sample categories and confidence. Through cascading and cooperative work of the content awareness stripe pyramid and the deformable attention area proposal network, the detection precision and robustness of the multi-scale target in the remote sensing image are effectively improved.

The invention provides a multi-scale pyramid weighted fusion underwater image enhancement method based on double prior, and the method comprises the steps: obtaining a degraded underwater image, and carrying out the global and local cooperation body color calibration of the degraded underwater image; decomposing the color correction image into a base layer, a detail layer and a noise layer by adopting a variational decomposition algorithm; performing spectral prior and transmissivity loss constraint on the base layer image to obtain a defogged image; fusing the detail layer image and the noise layer image to obtain a filtered image; performing enhancement processing on the filtered image by adopting a nonlinear mapping and contrast enhancement strategy to obtain an enhanced image; performing multi-level feature integration and reconstruction on the defogged image and the enhanced image by adopting a multi-scale pyramid adaptive weighted fusion method to obtain an underwater image with natural color and high visual definition; according to the method, the traditional image processing and variational optimization thought is combined to effectively correct the color deviation of the degraded underwater image, the image contrast and the detail definition are improved, and the visualization effect of the underwater image is improved.

The invention discloses a ship bollard identification method and system based on adaptive multi-scale multi-grid division, and belongs to the technical field of computer vision and image processing. The system divides an image foreground region and a background region through a visual saliency calculation module, carries out dense sampling in the foreground region and sparse sampling in the background region by adopting a non-uniform grid generation module, screens candidate grids in combination with gradient direction consistency and texture features, fuses candidate frames of different scales through a multi-scale image pyramid, and finally obtains a multi-scale image. And accurate positioning of the bollard is realized through the fine grid accurate positioning module. The method solves the problem that the prior art is insufficient in adaptability to ship size, shooting distance and resolution change, improves the precision and generalization ability of bollard positioning in a complex scene, and is suitable for bollard detection scenes of various ship images.

The invention belongs to the technical field of medical image processing, and particularly relates to a linear Transform general focus recognition method based on multiple perception and context guidance, and the method comprises the steps: extracting the multi-scale features of a medical CT image through a backbone network, obtaining the edge gradient features in parallel, making up the missing of focus boundary information through an edge perception feature enhancement module, and carrying out the recognition of the focus. And then global feature modeling under linear complexity is realized through a polarity perception feature interaction module, a high-discrimination-force multi-scale feature map is generated by using a context-guided feature pyramid network, and finally the model is optimized by combining a Hungary algorithm and a joint loss function based on an end-to-end detection architecture of set prediction. The problems that the focus boundary is fuzzy, feature interaction and calculation efficiency are balanced, and the focus and background separation degree is weak are effectively solved, double improvement of calculation efficiency and detection precision on massive medical image data is achieved, and reliable support is provided for clinical precise auxiliary diagnosis.

The invention provides a medical image registration method and equipment. The method is applied to the technical field of medical image processing. The method comprises the steps that a to-be-registered image pair used for medical image registration is acquired, the to-be-registered image pair comprises a floating image and a fixed image, the floating image is an image needing spatial transformation in the registration operation, and the fixed image is a reference standard correspondingly consistent with the spatial position and feature of the floating image in the registration operation; performing hierarchical feature extraction on the floating image and the fixed image through a registration model to obtain a first multi-scale feature pyramid and a second multi-scale feature pyramid, and performing correlation perception registration on multi-scale features in the first multi-scale feature pyramid and the second multi-scale feature pyramid to obtain deformation field data; and performing spatial transformation on the floating image based on the deformation field data to obtain a target registration image. According to the invention, efficient and accurate medical image registration is realized.

The invention belongs to the technical field of medical image processing, and particularly relates to a prostate MRI-TRUS deformable image registration method based on structure perception decoupling learning. The method comprises the following steps: designing an anatomical maintenance intensity disturbance module, simulating intensity and artifact differences between different modes on the premise of keeping an anatomical structure unchanged, and generating diversified appearance samples to improve the adaptability of the model to mode changes; a double-flow encoder structure is constructed, space attention latent consistency loss is introduced into a multi-layer feature space, the structure consistency is restrained from the feature level, and deep structure representation learning with the unchanged appearance is achieved; and an enhanced pyramid decoder is adopted to fuse multi-scale structural features layer by layer to predict a registration deformation field, so that high-precision cross-modal alignment is realized. According to the method, structural consistency constraint is carried out on the hidden space level, so that the influence of modal difference and artifacts can be effectively reduced, and the registration precision is improved.

The invention relates to the technical field of computer vision and image processing, and particularly discloses an unmanned aerial vehicle identification and detection method under a target part feature missing condition. The method comprises the following steps: (1) making an initial data set by adopting aerial pictures of an unmanned aerial vehicle; (2) performing labeling and data enhancement processing on the initial data set to obtain a training set and a verification set; (3) a YOLOv8 target detection model is improved, a standard convolution module (Conv) of a backbone network (Backbone) is replaced by dynamic deformable convolution (DEConv), a SimAM attention mechanism is introduced behind a last C2f module of the backbone network and in front of a spatial pyramid pooling layer (SPPF), and in a neck network (Neck), the C2f module is replaced by a C2f-SimAM module, and the standard convolution module is replaced by the dynamic deformable convolution; an additional branch for capturing key features is added in an output branch of a detection head (Head). According to the method, the recognition and detection capability of a target with partial feature missing can be remarkably improved.

The invention discloses a monitoring image multi-target tracking method based on deep learning, and relates to the technical field of digital image processing, and the method comprises the following steps: S1, generating a multi-scale feature pyramid; s2, extracting causal feature vectors; s3, generating a discrete codebook index; s4, constructing a dynamic graph; s5, inputting the dynamic graph into an improved ASTGCN network, iteratively fusing space-time attention weights and Hamilton dynamics evolution characteristics between nodes through cascaded Hamilton space-time blocks, and generating a final prediction state of each historical track node; s6, solving an optimal correlation matching matrix by using a Hungary algorithm; and S7, extracting all active track information. According to the method, the limitations of identity drift caused by variable appearance characteristics and inaccurate prediction caused by lack of physical priori in a traditional multi-target tracking method are overcome, and an efficient, accurate and robust solution is provided for intelligent video monitoring.

The invention provides a multi-scale context aggregation and dynamic supervision medical image segmentation method and application thereof, and belongs to the technical field of medical image processing. In order to solve the problems of weak non-linear feature fitting ability, global context missing and unstable training convergence in the prior art, the ResUKAN + network is constructed. According to the method, a residual KAN convolution module is embedded in a full level of an encoder, and nonlinear feature extraction is enhanced by using a B-spline function; a multi-scale context aggregation module is arranged on a bottleneck layer, and dynamic pyramid pooling and a double attention mechanism are fused to capture global dependency; a dynamic auxiliary supervision head is introduced at the tail end of a decoder, complementary features are extracted through a heterogeneous receptive field, and loss calculation is optimized in combination with a dynamic weight mechanism which is exponentially attenuated along with a training period. According to the method, the segmentation precision and robustness of the fuzzy boundary and the multi-scale focus are remarkably improved, and the method is suitable for medical image intelligent diagnosis.

The invention relates to a lightweight multi-scene disease and pest detection method and system based on RT-DETR. According to the scheme, firstly, a standardized image processing link is constructed, multi-band feature extraction is carried out on an input image by using a convolutional backbone network introduced with wavelet transform, an effective receptive field is expanded in a frequency domain through wavelet decomposition and an inverse reconstruction mechanism, and feature capture of a tiny insect pest target is enhanced while calculation redundancy is reduced. Furthermore, a bidirectional feature pyramid network including global and local double-branch collaborative modeling is adopted, cross-level dynamic interaction and gating fusion are performed on multi-scale features, and environmental noise interference such as veins and illumination under a complex farmland background is effectively inhibited. And finally, establishing a homography mapping model from a pixel plane to a geographic space according to camera calibration parameters, converting a visual detection result into a spatial distribution diagram layer with latitude and longitude information, and realizing dimension crossing of pest and disease damage monitoring from single-point identification to region-level risk assessment.

The invention discloses a skin lesion segmentation method and system based on dual-path perception multi-stage fusion, and relates to the technical field of medical image processing, and the method comprises the steps: carrying out the block division of a skin image, inputting into an encoder, and filtering a key lesion region through employing a region sparse attention mechanism; performing down-sampling on the feature map by using a block merging operation, and constructing a multi-scale feature pyramid; skin image features are processed by using a dual-path sensing multi-stage fusion network, deepest features generated by an encoder are processed by a spatial frequency dual-path cascade sensing module, and two-stage space-frequency collaborative optimization is executed; the method comprises the following steps: establishing jump connection between an encoder and a corresponding decoder through a space channel double-path parallel sensing module, recovering spatial resolution through block expansion operation in combination with modulated encoder characteristics, and generating a pixel-level lesion probability graph. According to the method, the context sensing capability across the space domain and the frequency domain is remarkably enhanced, and an accurate and detailed skin lesion segmentation result is obtained.

The invention relates to the technical field of image processing and intelligent identification, and discloses an image segmentation method for identifying a mineral boundary in a table ore zone. The method comprises the steps of constructing a multi-branch single-output enhancement unit based on gating enhancement and multi-head attention combined dynamic fusion, constructing a multi-scale fusion module of a split-level fusion pyramid structure based on multi-head attention and residual fusion, constructing a trunk module based on convolutional layer and enhancement unit stacking, and constructing a multi-scale fusion module of a split-level fusion pyramid structure based on multi-head attention and residual fusion. Constructing an image recognition segmentation model by combining step-by-step down-sampling, a multi-scale fusion module and a prediction head; and obtaining an ore zone image of a shaking table operation site of the reselection workshop, inputting the ore zone image into the image recognition segmentation model to obtain a recognition result, and completing shaking table separation of the mineral particles based on the recognition result. The problem of insufficient segmentation precision caused by particle size difference, illumination interference, fuzzy boundary and real-time requirement in the existing industrial mineral separation process is solved.

The invention relates to the technical field of image processing, in particular to a gland structure irregularity quantitative analysis method and system based on a gastroscope image, and the method comprises the steps: carrying out the color correction and image contrast enhancement of an obtained original gastroscope image, and constructing a corresponding image pyramid; inputting the images in the image pyramid into an improved deep learning model for learning by adopting a multi-channel fusion input mode, and obtaining a gland binary mask after morphological processing; based on three dimensions of morphology, structural arrangement and complexity, quantitative features of a single gland structure are extracted from the gland binary mask; the extracted multi-dimensional quantitative features are trained through a gradient lifting decision tree model, an irregularity index is output, feature importance analysis is carried out, and the gland structure in the gastroscope image can be accurately recognized and quantized.

The invention discloses a remote sensing image segmentation method and system based on a fine screening double-domain attention mechanism, and belongs to the technical field of remote sensing image processing. A multi-level feature extraction network with a fine screening double-domain attention unit as a core is constructed, and pyramid-like cross-layer residual connection is fused; semantic expression and spatial distribution perception capability in the remote sensing image are remarkably enhanced, a fine screening double-domain attention unit is introduced into each layer of coding module, a multi-layer feature fusion mechanism is matched, sufficient modeling of texture, structure and context information of the remote sensing image under different scales is achieved, and remote sensing image segmentation is completed on the basis.

The invention belongs to the field of image processing and computer vision, and relates to an object-level contrast learning method for multi-modal target detection. The invention provides an object-level intra-modal and cross-modal combined contrast learning method aiming at the problems that multi-modal remote sensing image labeling cost is high, a pre-training structure and a detection task are not matched, and complementary information among modals is insufficient in utilization. The method comprises the following steps: obtaining and preprocessing paired visible light and infrared images, and generating and screening candidate boxes; performing multi-view enhancement on the two-mode image and synchronously mapping a proposal box; a double-branch pre-training network is constructed, object features are extracted in a multi-level mode in a feature pyramid, and target network parameters are updated through intra-modal and cross-modal comparison loss joint optimization and index moving average. And after pre-training is completed, migration to a detection model is carried out, multi-modal fusion detection is realized through fine adjustment of a small amount of annotation data, the detection precision and robustness can be improved, and annotation dependence is reduced.

The invention discloses a foot wound surface detection method and system based on image recognition, and relates to the technical field of medical image processing and analysis, and the method comprises the steps: extracting foot wound surface features of high-resolution foot wound surface images under different scales through a multi-scale feature fusion network, constructing pyramid hierarchies, and obtaining a multi-scale feature fusion network; feature fusion is carried out from top to bottom through the feature pyramid structure, and a multi-scale foot wound feature map is generated; inputting the multi-scale foot wound feature map into a segmentation network for cross-modal interaction and fusion, obtaining a regional physiological characteristic map and a regional probability map by using a deformable convolutional network and a decoding network, and comparing the regional physiological characteristic map and the regional probability map through an adaptive threshold map to generate a binary mask foot wound image; according to the method, the segmentation accuracy is improved, and detailed wound surface information is also provided, so that the accuracy and efficiency of clinical diagnosis are greatly improved.

The application discloses an image feature extraction method, system and medium, and belongs to the field of image processing. In view of the problems of poor accuracy and complex operation of existing image feature point extraction, the application provides an image feature extraction method, which comprises the following steps: acquiring an image, constructing an image pyramid after carrying out gray processing on the image; calculating the contrast of the image; if the contrast of the image meets the requirement, directly performing the next step; if the contrast of the image does not meet the requirement, carrying out image enhancement processing on the image; extracting FAST corner points in the image and screening; and outputting feature points. The application guarantees the invariance of feature scales by constructing the image pyramid; subsequently, the image with texture is screened according to the contrast, and the operation of improving image details is performed on the image, so that the image feature points are more prominent; the whole method is simple in operation, high in work efficiency, improves the image feature point extraction effect, and further improves the image registration effect, thereby guaranteeing the positioning and navigation effect of a robot based on images.

The present application belongs to the technical field of material testing and visual detection, and particularly relates to a visual observation system for high-temperature and high-pressure testing of organic silicon sealant, which comprises a high-pressure sealing cavity, a transparent observation window, a high-frame-rate industrial camera, a multi-spectrum coaxial light source system, an image acquisition and control module, and an image processing and analysis module; multi-scale space-time features are extracted through construction of an image pyramid and a three-dimensional convolutional neural network, a constitutive model of the organic silicon sealant is embedded into a light flow calculation framework to reconstruct a deformation field and a stress field in accordance with physical laws, a space-time graph convolutional network is used to identify rupture, flow and debonding failure modes, a physical information neural network is used to predict failure time and an extension trajectory, and finally, a safety working condition window is output by fusing temperature and pressure data. The present application realizes intelligent, quantitative and accurate observation of the dynamic failure process of the organic silicon sealant under high-temperature and high-pressure working conditions, and provides reliable data support for sealant selection and injection molding process optimization.

The application relates to the technical field of image processing, in particular to an image super-resolution reconstruction method based on a multi-scale content perception mixer, which is realized by using an adaptive processing mechanism. The method comprises the following steps: shallow feature extraction is performed on a low-resolution image to be reconstructed, so as to obtain an initial shallow feature map; feature enhancement based on a feature pyramid and an attention mechanism is performed on the shallow feature map, so as to obtain a deep feature map; multi-scale content perception prediction is performed based on the deep feature map, so as to generate guide information for guiding calculation allocation, the guide information comprising a window classification binary mask and a window size; different image regions are allocated to different calculation paths for processing based on the guide information; the feature maps output by the calculation paths are recombined and fused, and then enlarged to a target resolution, so that a high-resolution image is finally obtained. The method realizes accurate classification of image regions and on-demand allocation of calculation resources, and significantly reduces the calculation complexity and the memory occupation.

The invention relates to the technical field of medical image processing, in particular to a PL medical image segmentation method combining deep supervision and a mixed loss function. The method comprises the following steps: carrying out cutting, registration, normalization, enhancement and dicing preprocessing on an input three-dimensional medical image; inputting the image blocks into an encoder-bottleneck layer-decoder network; an encoder extracts multi-scale features through pooling of a residual VGG block and a dual-channel spatial pyramid; performing depth feature abstraction and context modeling on the bottleneck layer; the decoder performs up-sampling through transposition convolution, fuses encoder features transmitted by jump connection, utilizes space attention to gate and focus a target, and optimizes the features through an ASPP module and a residual block; and setting output in a middle layer and a final layer of the network, and training the network through a mixed loss function combining Dice loss and cross entropy loss and a deep supervision mechanism. The precision and robustness of medical image segmentation can be effectively improved.

The invention discloses a multi-source heterogeneous satellite remote sensing image matching method based on sparse sampling description, and relates to the technical field of remote sensing image processing, and the method comprises the steps: generating an image pyramid of an input image, extracting a feature point set based on a convolution image of each layer of image, and carrying out the fusion to obtain a convolution image corresponding to a convolution direction; applying the constructed sparse descriptor at the feature points, determining the main direction of the feature points according to the statistical result of the convolution value of each pixel point in the sparse descriptor range, sampling at each sampling point of the sparse descriptor to obtain description vectors, and splicing the description vectors to obtain feature description vectors; matching is carried out based on the feature description vectors of the multiple input images to obtain a matching point pair set; the number of sampling points of the sparse descriptor is lower than that of the descriptor in a dense sampling mode in the related technology, the matching calculation amount can be further reduced, feature point matching between the images can be completed only through fewer sampling points, and the calculation efficiency can be improved.

The application discloses a tone mapping method and system based on a lookup table and a Laplace filter, and belongs to the technical field of image processing.The input image is first decomposed into an adaptive Laplace pyramid;then a low-frequency image at the bottom layer of the adaptive Laplace pyramid is input into a weight predictor to obtain a pixel-level weight map, and a three-dimensional lookup table is used to perform three-linear interpolation on the low-frequency image to obtain a preliminary mapping image; then the preliminary mapping image and the weight map are fused pixel by pixel to generate an image at the bottom layer of the fine-tuned Laplace pyramid; then a filter parameter prediction module is used to learn a parameter value map of the remaining layer images of the adaptive Laplace pyramid, and the parameter value map is applied to a local Laplace filter to obtain the remaining layer images of the fine-tuned Laplace pyramid; finally, the fine-tuned Laplace pyramid is reconstructed to obtain a tone-mapped image.The application can retain local edge details of an image while performing global tone mapping.

The application belongs to the technical field of medical image processing, and particularly relates to a pancreas image segmentation method based on complementary attention; the method comprises the following steps: acquiring a pancreas CT image, pre-processing the pancreas CT image, extracting features of the pre-processed pancreas CT image by using a residual dense module to obtain a first feature map; processing the first feature map by using a progressive pyramid pooling module to obtain a second feature map; processing the second feature map by using a main branch decoder and an edge branch decoder respectively to obtain a region feature map and an edge feature map; processing the region feature map and the edge feature map respectively by using a complementary attention mechanism to obtain a pancreas result map and an edge result map; and the application can solve the problems of over-segmentation and under-segmentation, thereby improving the accuracy of pancreas segmentation.

The present application relates to the technical field of die cutting detection, and more particularly to a die cutting appearance defect detection method and system based on image processing, which comprises: acquiring an image and counting a global gradient main direction, inputting a feature extraction network containing a strip convolution group, correcting gate weights based on the global main direction bias, and adaptively adjusting the size of the deep convolution kernel according to the weight distribution entropy value to accurately extract the strip features; when the features in the pyramid are fused across layers, the high-layer gradient amplitude is used to perform pixel-level denoising screening on the low-layer features; finally, the direction weighting of the bounding box regression loss is performed based on the main axis direction of the defect, and the non-maximum suppression threshold is anisotropically relaxed. The present application effectively filters out complex background noise and improves the accuracy of industrial die cutting detection.

The invention discloses a bulk electron microscopic image splicing method based on a multi-scale image pyramid, and belongs to the technical field of image processing. According to the method, two-stage local abnormal value filtering and RANSAC enhancement based on local geometric constraints are introduced in the image registration stage, and image enhancement, feature re-extraction and multi-iteration TPS displacement field calculation are adopted in local deformation correction, so that the adaptability and robustness of the algorithm to a complex biological sample microstructure and potential deformation are jointly improved. By introducing a multi-scale image pyramid structure and a hierarchical processing strategy, a complex fusion problem of a high-resolution image is decomposed into a plurality of low-resolution sub-problems, and the calculation efficiency and the memory utilization rate are remarkably improved; and the problem of fuzzy texture details caused by a traditional weighted average method is effectively avoided.

The invention relates to the technical field of underwater image processing, in particular to an underwater image enhancement method based on double-branch complementary input and cross-branch cross-layer state transition, and the method comprises the steps: obtaining an underwater image; an underwater image enhancement network model is constructed, the underwater image enhancement network model comprises a first branch, a second branch and a state transition module, the first branch comprises a self-adaptive feature extraction module, a multi-scale spatial pyramid pooling module, a channel splicing fusion layer and a tail end self-adaptive feature extraction module, and the second branch comprises a multi-scale spatial pyramid pooling module. The second branch comprises a channel intensity inversion layer, a self-adaptive feature extraction module, a multi-scale spatial pyramid pooling module, a channel splicing fusion layer and a tail end self-adaptive feature extraction module; training an underwater image enhancement network model; and inputting the test set into the trained model, and outputting an enhanced underwater image. According to the method, the color recovery precision and the detail retention performance of the underwater image in a variable environment are improved.

The invention discloses a waterlogging intelligent identification method and related equipment suitable for an urban low-illumination environment, and the method comprises the steps: carrying out the illumination normalization processing of a low-illumination image, obtaining a normalized output image, extracting a multi-layer feature through a multi-layer gradual downsampling operation through a feature extraction module, and obtaining an intermediate feature map; decomposing the intermediate feature map into content features and style features, and further removing the style features from the content features by minimizing an orthogonal loss function to obtain decoupled content features; performing feature enhancement on the decoupled content features to obtain enhanced content features; based on the enhanced content features, capturing multi-scale context information through parallel convolution by using a cavity space pyramid module to obtain target features; and based on the target features, a decoder is used to carry out progressive up-sampling to output a segmentation mask, and a waterlogging identification result is obtained. The waterlogging identification accuracy in the low-illumination scene can be improved, and the method can be widely applied to the technical field of image processing.

The application relates to the technical field of image processing, and provides an industrial visual inspection method based on a deep convolutional neural network and related devices. The method comprises the following steps: acquiring an original image of an industrial product to be inspected, and performing environment interference suppression processing on the original image to obtain a pretreated image; performing multi-scale feature extraction on the pretreated image based on a preset convolutional neural network to obtain multi-scale local feature information; performing global analysis on the multi-scale local feature information based on a preset Transformer encoder to obtain global feature information; performing weighted fusion on the multi-scale local feature information based on a preset attention gate mechanism, taking the global feature information as a guide, to obtain a multi-scale attention-enhanced feature pyramid; and performing analysis on the multi-scale attention-enhanced feature pyramid based on a preset parallel prediction subnetwork to generate category and position information of defects in the original image. The method is helpful to improve the detection accuracy of micro defects.

The application provides a kind of intelligent enhancement and feature processing method for film and television image, it is related to image processing field, its steps include: by constructing lightweight adaptive convolution module, the cross-scale feature of single frame film and television image is decomposed, and multi-resolution pyramid level is formed, and the temporal correlation between adjacent frames is combined, multi-scale features are reconstructed and unified feature space fusion Feature flow, obtain film and television image feature representation;Introduce the feature level adaptive enhancement mechanism of time sequence perception, based on multi-scale structural saliency information and interframe temporal consistency constraint, the dynamic selective strengthening and redundancy suppression of film and television image feature representation are carried out, and the enhanced film and television image feature representation is obtained;Construct feature-driven adaptive reconstruction unit, according to the enhanced film and television image feature representation, the content-aware pixel-level reconstruction and detail compensation of single frame film and television image are carried out, and enhanced film and television image is generated.

The invention relates to the technical field of image processing, in particular to an electric vehicle charging port identification method and system based on deep learning. The method comprises the steps of obtaining an original image of an automobile charging port and performing preprocessing to obtain an initial processing image; for any pixel point, calculating the diffuse reflection confidence of the pixel point; weighting the gradient amplitudes of the pixel points based on the diffuse reflection confidence coefficient to obtain an optimal HOG (Histogram of Oriented Gradient) feature; performing sliding search on each level of the pyramid of the initially processed image by using a sliding window, and calculating the charging port aggregation degree of any sliding window; adjusting a preset basic deformation weight in a score function of the DPM model based on the charging port polymerization degree to obtain a self-adaptive deformation penalty weight corresponding to the sliding window; and calculating a comprehensive score of each sliding window by using a DPM model score function based on the adaptive deformation penalty weight, and determining the position of the charging port based on the comprehensive score. The charging port identification accuracy can be improved.