91 results about "Class activation mapping" patented technology

The invention discloses a remote sensing image pseudo label generation method based on prototype perception learning, and a weak supervision change detection model training and detection method, and belongs to the field of remote sensing image change detection. The invention aims to solve the problem that a category activation graph is incomplete in existing remote sensing image weak supervision change detection and the problem that an obtained detection model is unstable in change recognition. According to the method, feature maps of different scales before and after change are extracted from remote sensing images before and after change through a pseudo label generation model, change features are obtained, a class activation mapping map and a class activation mapping map of an unchanged class are obtained, and the class activation mapping map is segmented into response area masks through a threshold value and used for performing weighted aggregation on fusion features to obtain a class prototype; generating an enhanced class activation mapping graph based on the fusion feature and the class prototype, and selecting the class with the highest similarity as a pseudo tag of the pixel; and training based on the pseudo label data of the pixels to obtain a change detection model, wherein the model is used for remote sensing image change detection.

The invention provides a seawater buoyant raft cultivation information extraction method based on an interpretable polarization deep learning network, and belongs to the technical field of ocean remote sensing and artificial intelligence crossing. Wherein the interpretable polarization deep learning network comprises three key components: a polarization scattering type activation mapping module, an interpretable segmentation consistency learning module and a refined type activation and edge guiding network. The polarization scattering class activation mapping module generates a high-quality class activation mapping graph by combining polarization scattering characteristics and gradient weights of the fully-polarized SAR image; the interpretable segmentation consistency learning module uses the generated class activation mapping graph to guide the model to pay attention to the characteristics of the breeding area in the training process; the refined class activation and edge guidance network adopts a double-branch architecture, semantic segmentation and edge supervision are performed respectively, and an accurate segmentation result is generated. According to the method, background misjudgment can be remarkably reduced, and a clear and detailed segmentation result is ensured; and the method has relatively high applicability.

The invention discloses an interpretability analysis and decision sharing verification method and system for a rectal cancer prognosis model, and relates to the field of medical artificial intelligence interpretability. The method comprises the following steps: carrying out gradient weighting class activation mapping analysis on a prognosis model to generate an image thermodynamic diagram; calculating the contribution degree of the multi-modal features by using an SHAP interpreter; an integrated visual interface is constructed, and patient data, model prediction and the explanation result are presented to a doctor together; the doctor performs independent risk assessment based on the interface information; finally, decisions of doctors and the model are compared, and model auxiliary efficiency is evaluated. Through a doctor-model decision sharing verification mechanism which is explained and innovated in a multi-level mode, the transparency and clinical credibility of the complex AI prognosis model are remarkably improved, the value of time sequence data in dynamic risk assessment can be verified, and clinical landing application of the AI model is powerfully promoted.

The invention belongs to the technical field of rehabilitation assessment, and provides a cerebral apoplexy rehabilitation assessment method and system based on multi-mode electroencephalogram and myoelectricity fusion. According to the method, 59-channel electroencephalogram signals and 14-channel electromyographic signals are preprocessed, mutual information between channels is calculated to construct a correlation matrix, spatial-temporal features are extracted in combination with a convolutional neural network (CNN) and a Transform self-attention mechanism, and rehabilitation level classification is achieved. According to the method, coherence between electroencephalogram / myoelectricity channels is quantified by adopting mutual information, and 59 * 59 and 14 * 14 dimensional feature matrixes are constructed; a CNN-Transform hybrid model is designed to optimize feature fusion, and the classification precision is improved; gradient weighted class activation mapping (Grad-CAM) is introduced to analyze feature saliency, and correlation between an electroencephalogram channel and a focus is revealed; and developing a visual evaluation system, and displaying the multi-modal data and the historical trend in real time. The system stores patient information and evaluation results through a MySQL database, and supports doctor-patient online interaction. Compared with traditional scale evaluation, the method has the advantages that the evaluation objectivity is improved by using multi-modal objective data, the model credibility is enhanced by combining interpretability analysis, and technical support is provided for formulating a personalized rehabilitation scheme.

The invention discloses a semiconductor wafer defect detection method, device and equipment and a storage medium, and the method comprises the steps: preprocessing an obtained to-be-detected image of a semiconductor wafer, and obtaining an input image corresponding to the to-be-detected image; inputting the input image into the classification model to obtain all target defect categories reaching a confidence threshold and confidence values; generating a thermodynamic diagram corresponding to each target defect category based on a category activation mapping mode; respectively processing all the thermodynamic diagrams to obtain an effective area with defects on the input image; target areas on the to-be-detected image are determined based on the effective areas, and after target defect categories and confidence values corresponding to the effective areas are marked for each target area, the marked to-be-detected image is output. According to the method, the target detection effect is achieved through the classification model and the class activation mapping mode to replace a target detection algorithm, so that the target detection algorithm does not need to be trained by tedious data labeling, and the manual labeling cost is reduced.

The invention relates to the technical field of fault diagnosis, in particular to a few-label self-supervised learning fault diagnosis method based on an interpretable neural network, and the method comprises the steps: collecting time domain data of a sensor of an aero-engine under different health conditions, carrying out the standardization processing, and dividing the time domain data into a pre-training set, a training set, a verification set and a test set; performing fast Fourier transform and data enhancement on the time domain data in the pre-training set to obtain frequency domain data and enhanced time domain data; constructing a pre-training framework, and performing pre-training to convergence based on the frequency domain data and the enhanced time domain data to obtain a pre-trained time encoder; constructing a fault diagnosis model based on the pre-trained time encoder, and performing iterative training to convergence by using the training set to obtain a trained fault diagnosis model; and performing fault diagnosis on the test set by using the trained fault diagnosis model, and performing visual interpretation on the diagnosis process of the fault diagnosis model by using a gradient weighting class activation mapping technology.

The invention relates to the technical field of artificial intelligence, in particular to a class activation mapping interpretability method fusing a spatial perturbation mechanism, which comprises the following steps: a first stage: capturing contribution of each channel to prediction, and obtaining a channel attention weight of a deep learning model through a gradient method; in the second stage, disturbance analysis is introduced, and space importance weights given to image samples by the model are obtained; and a third stage: fusing the interpretation information of the two view angles through weighted integration, and generating an interpretation saliency map of the region-level granularity. According to the invention, the dual positioning capability of an interpretable technology in a channel domain and a space domain can be improved.

The invention discloses a medical image automatic identification system based on a neural network, and relates to the technical field of medical image identification. The method is used for solving the problem that early recognition of neurodegenerative diseases is difficult due to medical image and genome data splitting and poor model interpretability in the prior art. The method comprises the following steps: firstly, extracting multi-scale features of a brain structure through a three-dimensional convolutional neural network and a self-attention mechanism, calculating a multi-gene risk score based on a risk site, and encoding the score into a feature vector; secondly, using a cross attention mechanism to take gene features as query vectors, fusing the gene features with image features, and generating brain structure anomaly features under gene regulation; then, gradient weighting class activation mapping is applied to generate a visual thermodynamic diagram, and gene-image association weight weighting is combined to construct a brain region risk distribution diagram; and finally, a high-risk brain region space coordinate set is extracted through threshold segmentation, and an accurate quantification basis is provided for early recognition.

The invention discloses an SAR image weak supervision semantic segmentation method based on decision optimization pseudo label generation, and the method comprises the following steps: 1, training a multi-label classification network, and obtaining class activation mapping results of different layers of a neural network and pixel-level prediction pseudo labels; 2, fusing different levels of class activation mapping results to obtain a class activation mapping result with higher confidence; step 3, forming a decision by utilizing the relation between different layers of pseudo labels on the pixels, and extracting multi-layer decision pixel relation information; 4, constructing a relation network among pixels, and obtaining a centroid displacement vector and boundary features of each pixel in the image; 5, training a relation network between pixels by using the extracted pixel relation information to obtain a pixel semantic affinity matrix; step 6, performing pseudo tag optimization by using the pixel semantic affinity matrix; and step 7, training a semantic segmentation network by using the optimized pixel-level pseudo labels, and completing SAR image weak supervision semantic segmentation. According to the method provided by the invention, the high-quality polarized SAR image pixel-level pseudo tag can be generated, and pixel-level semantic segmentation of the SAR image under the condition of image-level annotation weak supervision is realized.

The invention relates to a pooling and up-sampling fusion CAM method based on information entropy and an attention mechanism. The method comprises the steps of image preprocessing, feature extraction, information entropy calculation, attention weight matrix design, weighted feature graph and pooling, up-sampling and residual connection, category weight assignment, CAM generation and the like. By introducing the information entropy theory, the complexity and importance of information in the feature map can be accurately evaluated, attention weight design is guided, key information is focused, redundancy is suppressed, and the capability of processing complex images by the model is remarkably improved. And in combination with an attention mechanism, feature representation is more effective and accurate. Through pooling and up-sampling technology fusion, the size of a feature map is flexibly adjusted, key features are reserved, and subsequent tasks are facilitated. Through the application of global average pooling and a pre-training classifier, a class activation mapping graph is generated, the performance and accuracy of a computer vision task are improved, and the explanatory performance of a model on picture processing is improved.

The invention provides a diffusion noise enhancement method for a long-tail remote sensing image. According to the method, before training, a specially designed condition prompt set and an original training set are used for jointly guiding a diffusion model to generate a new sample, and a CLIP model is used for screening. In addition, the invention provides two effective generation data utilization strategies, and optimization is respectively carried out at the end of the first-stage training and in the second-stage training process. Firstly, a DiffCam-Mix module is designed, the module uses a Grad-CAM + + class activation mapping method to extract a background of a generated image and a foreground of an original image, the background and the foreground are mixed, and mixed data with real features of the original image and diversity of the generated image are constructed. And secondly, in a training process, introducing a comparative learning method based on cosine similarity, so that the mixed data and the corresponding original data are kept consistent in a feature space, thereby further calibrating data distribution. The method is suitable for a long-tail distribution scene in a remote sensing image classification task, the number of tail category samples can be effectively increased, the classification precision of the model is improved, and the recognition capability of long-tail data is improved.

The invention discloses a hospital terminal threat behavior recognition method based on interpretable deep learning, and particularly relates to the technical field of network security, threat intelligence analysis and interpretable artificial intelligence. According to the method, a multi-modal hospital terminal operation data set fusing time sequence images, logs and other data is constructed, and different modal data are effectively integrated by adopting a multi-modal data fusion method; integrating a dynamic attention mechanism into a long short-term memory network model for adaptively extracting key features in a terminal user behavior sequence; and a double-layer interpretable framework based on gradient weighted class activation mapping and a local interpretable model irrelevant interpretation technology is introduced, so that high-precision recognition and interpretable analysis of the threat behavior of the hospital terminal user are realized. According to the method, complex threat behaviors in the hospital terminal can be efficiently and accurately identified, the limitation of a traditional method in the aspects of data diversity, interpretability and hidden threat detection is overcome, and the transparency and efficiency of threat analysis are enhanced.

The invention discloses a method and device for predicting the transmembrane capability of a substance molecule, terminal equipment and a storage medium, and the method comprises the steps: obtaining a target molecular substance, and determining a molecular structure diagram corresponding to the target molecular substance; according to a pre-trained prediction model, determining a permeation free energy prediction value of the target molecular substance; according to the prediction model, multilevel feature extraction is carried out on a molecular structure diagram of sample substance molecules by utilizing a graph neural network model, contribution values of atoms in the sample substance molecules to transmembrane behaviors are evaluated through a gradient weighted class activation mapping method, and permeation free energy prediction values corresponding to the sample substance molecules are output. The graph neural network is trained to obtain the prediction model, and the prediction model is adopted to predict the transmembrane capability of the target molecular substance, so that intelligent screening of the transmembrane capability of a large-scale drug molecular library can be realized, the prediction efficiency and flux are remarkably improved, and intelligent prediction of the transmembrane free energy of substance small molecules is realized.

The invention discloses a fault diagnosis method, system, equipment and medium for a wind power gear box, and relates to the technical field of fault detection, and the method comprises the steps: collecting a vibration signal of the wind power gear box, and obtaining a time-frequency diagram of the vibration signal; capsule feature tensors in the time-frequency graph are obtained, a time-frequency two-dimensional network is formed in space, and corresponding query tensors are extracted from a time axis and a frequency axis of the time-frequency two-dimensional network by adopting different convolution kernels; dot product matching is carried out on the corresponding query tensors and a preset shared key value pair, attention distribution of each query tensor is obtained, and aggregation output in the directions of the time axis and the frequency axis is fused through the attention distribution; and performing class activation mapping on the aggregated output, determining a fault category of the wind power gear box, generating a class activation graph focused on the key time-frequency region through the fault category, and displaying a discrimination basis. The method can focus on a key time-frequency region with clear physical significance, and obviously improves the discrimination transparency and engineering credibility of the model.

The invention discloses a pathological image classification method based on multi-stage dynamic collaborative optimization, and relates to the field of medical image processing. The method is innovatively characterized in that firstly, dynamic semantic pseudo package generation is carried out, and the problems of semantic confusion and parent package feature distribution loss caused by a traditional pseudo package division mode are solved; 2, a dynamic sparse graph attention expert network: by constructing a dynamic sparse graph structure and combining a multi-head self-attention and expert network, focus features are focused and redundancy calculation is inhibited; the uncertainty driving key sample mining module is used for mining uncertain samples near decision boundaries by utilizing cross-multi-head feature fusion calculation class activation mapping and improving the classification capability of the model on early lesions; and 4, according to a dynamic weight collaborative optimization strategy, dynamic weighting is carried out through a dual-stage loss function, and the model potential is further mined while the training stability is ensured. Through multi-stage collaborative optimization, the classification accuracy and the clinical diagnosis efficiency are effectively improved, and an efficient method is provided for complex pathology analysis.

The invention discloses a model decision interpretability method fusing integral gradient and class activation mapping, and belongs to the technical field of deep learning interpretability. Aiming at the problems of gradient noise interference, insufficient space positioning and the like existing in an existing single interpretability method, the method is optimized through five key steps: firstly, extracting a feature map set of the last convolutional layer of a deep learning model; secondly, calculating an integral gradient of the feature map to a target category based on a path integral idea; thirdly, obtaining a feature map weight through global average pooling; then, weighted summation is carried out, and an initial attribution thermodynamic diagram is generated through ReLU activation; and finally, a high-resolution thermodynamic diagram is obtained in combination with guided gradient optimization. According to the method, gradient stability of Integrating Gradients and spatial positioning advantages of Grad-CAM are fused, invariance and sensitivity axioms are realized, model adaptability is maintained, accuracy and robustness of interpretation results are remarkably improved, and the method is suitable for key fields such as signal processing and image classification.

The invention discloses a multi-modal health knowledge generation and integration system, which is applied to the technical field of fusion and application of health data, and comprises a multi-modal data acquisition and preprocessing module used for acquiring multi-modal health data and preprocessing the multi-modal health data; the multi-level fusion framework design sub-module comprises data-level fusion based on a time alignment algorithm, feature-level fusion based on a Transform cross-modal attention mechanism and decision-making-level fusion based on an ensemble learning algorithm; and the intelligent knowledge generation sub-module comprises health knowledge graph construction based on a graph neural network, personalized health risk report and intervention scheme generation based on a conditional generative adversarial network, visual image feature contribution degree based on gradient weighted class activation mapping and text data key factor analysis based on an SHAP value. According to the method, the bottleneck of a data island is fundamentally broken through, and the generation quality and the application value of health knowledge are improved.

The invention discloses a classification model defense method based on dynamic adversarial training and attention area denoising, and aims to improve the robustness of a classification model under adversarial attack. In the training stage, FGSM, PGD and Camp are dynamically selected according to the number of training rounds; w and other different confrontation sample generation methods are adopted, so that specific attacks of model over-fitting are avoided, and the generalization ability to various attacks is enhanced. Meanwhile, a CAM (Class Activation Mapping) is used for positioning an input image attention area, wavelet denoising processing is adopted, weighted fusion with an original image is carried out, countermeasure disturbance is accurately weakened, and key information is reserved. Experiments show that on a CIFAR-10 data set, the method significantly improves the model performance in FGSM, PGD and Camp; compared with a traditional method, the robustness accuracy is improved by 10%-20%, and the method can be widely applied to the fields of automatic driving, security and protection monitoring and the like.

The invention provides an improved multi-label image classification method based on multi-scale double dynamic graph convolution. The invention discloses the multi-label image classification method based on the improved multi-scale double dynamic graph convolution. The invention provides an improved multi-scale double dynamic graph convolution multi-label image classification method for solving the problems that traditional class activation mapping contains noise and the convolution effect becomes poor due to the high expansion rate of cavity convolution. The method comprises the following steps: firstly, introducing depth separable convolution to replace cavity convolution, and redesigning a combination mode of different scale features to obtain improved multi-scale features; secondly, a weight suppression mechanism is introduced to weaken class activation of irrelevant classes, and the class activation is multiplied by the improved multi-scale features to generate a new content awareness vector; and finally, inputting the content perception vector into a double-graph convolution fusion embedded network for final classification. Experimental results show that the precision of the provided model is improved on MS-COCO and VOC2007 data sets, and it is proved that the improved model has better performance.

This invention relates to a heatmap evaluation method and system based on class activation mapping and dynamic weights, comprising: acquiring crop images; inputting the crop images into a pre-trained deep learning model to output crop detection results; processing the crop detection results into a heatmap based on class activation mapping, and performing noise reduction processing on the heatmap to obtain a heatmap without background interference; performing color space conversion on the heatmap without background interference, and calculating the color difference between each pixel and a reference color, wherein the reference color includes a preset reference color and corresponding weight based on the heatmap; assigning a weight value to each pixel based on the color difference between each pixel and the reference color, and calculating an average temperature weight value; and completing a quantitative evaluation of the heatmap based on the average temperature weight value. This invention provides a new interpretive dimension and quantitative standard for the application of deep learning models in agriculture.

The present invention provides a method for extracting seawater raft aquaculture information from an interpretable polarimetric deep learning network, belonging to the cross - technical field of marine remote sensing and artificial intelligence. The interpretable polarimetric deep learning network includes three key components: a polarimetric scattering class activation mapping module, an interpretable segmentation consistency learning module, and a refined class activation and edge guidance network. The polarimetric scattering class activation mapping module generates a high - quality class activation mapping diagram by combining the polarimetric scattering characteristics of the fully polarimetric SAR image with the gradient weight; the interpretable segmentation consistency learning module uses the generated class activation mapping diagram to guide the model to focus on the characteristics of the aquaculture area during the training process; the refined class activation and edge guidance network adopts a dual - branch structure to perform semantic segmentation and edge supervision respectively, generating accurate segmentation results. The present invention can significantly reduce the occurrence of background misjudgment, ensure that the segmentation results are clear and detailed, and has strong applicability.

The invention belongs to the technical field of medical image processing and artificial intelligence, and particularly relates to a weakly supervised brain tumor segmentation method which comprises the following steps: acquiring a low-level glioma segmentation data set of a cancer genome map, and randomly dividing the data set into a training set and a test set according to patient division; based on the image data in the data set, carrying out transfer learning by using a pre-trained ResNet50 as a backbone network; a global optimization mechanism of an SZIO algorithm is utilized to perform adaptive optimization and intelligent screening on features extracted based on a traditional class activation mapping technology, and fine fusion and semantic enhancement of multi-level features are realized; and performing fine adjustment on the ResNet50 backbone network at a preset initial learning rate. According to the method, the performance breakthrough of weakly supervised brain tumor segmentation can be realized, the practicability, robustness and clinical transformation potential are relatively high, and an efficient and reliable intelligent segmentation solution is provided for the field of medical image analysis.

This invention discloses a deep learning-based weight estimation method, relating to the field of computer vision technology. By collecting side-view depth image data of a single sheep, and using sheep detection and sheep body semantic segmentation models to augment the collected data, a LiteHRNet network model is used as the backbone network of the weight estimation model to obtain semantic information at different levels from the sheep depth image data. Class activation mapping is used to visualize different head attention regions for weight estimation. Finally, a weight estimation model with small estimation error and concentrated sheep region attention is selected. This method can efficiently estimate sheep weight, thereby saving time and labor costs for livestock workers, and provides a research direction in deep learning for weight estimation of other similar animals.

The application discloses a method and system for predicting protein drug binding sites based on a multi-modal dynamic graph, comprising: obtaining amino acid sequence data and three-dimensional structure data of a protein, and generating evolutionary conservation features, structure graph topology features and sequence features respectively; inputting the features into a multi-modal fusion encoder to generate first fusion features; inputting the first fusion features into a prediction decoder to generate initial prediction probabilities; iteratively updating the structure graph topology features according to a preset three-graph update rule according to the initial prediction probabilities and residue dynamic communication scores; re-inputting the updated structure graph topology features into the multi-modal fusion encoder and the prediction decoder, and repeatedly executing until a preset iteration number is reached, to generate final prediction probabilities; and generating a residue importance heat map and outputting a prediction report based on the final prediction probabilities through a gradient weighted class activation mapping algorithm. The application realizes the cooperative optimization of prediction and graph structure, and significantly improves the accuracy of binding site prediction.

The invention discloses a car insurance claim settlement method and system based on a multi-modal large model, and the method comprises the steps: collecting the multi-modal data of a historical car insurance claim settlement case, employing a large-scale pre-training image-text large model to carry out the multi-task joint training and field adaptation fine tuning, and achieving the joint semantic understanding of an accident record text and a vehicle damage picture; the method comprises the following steps of: analyzing VIN (Vehicle Identification Number), embedding structured information, introducing a part association graph as an external knowledge constraint, and completing efficient fine adjustment of a model by adopting a hierarchical parameter freezing and low-rank adaptation technology under the condition of a small number of samples; in the reasoning stage, a structured damage analysis report, a compensation amount predicted value and confidence evaluation are output, a damage area thermodynamic diagram is generated in combination with a gradient weighting class activation mapping technology, and a visual decision basis is provided. The technical problems that traditional artificial claim checking is low in efficiency, high in cost and large in result difference, and an existing AI technology is high in data threshold, poor in field adaptability and lack of interpretability are solved.

The present invention discloses a method and terminal device for evaluating adversarial robustness based on class activation maps. The method comprises adding perturbations to original samples using an adversarial attack algorithm to obtain adversarial samples; obtaining class activation maps of a deep learning model based on the original samples and class activation maps based on the adversarial samples; performing adversarial defense training on the deep learning model using the original samples and the adversarial samples to obtain a trained deep learning model; obtaining class activation maps of the trained deep learning model based on the original samples and class activation maps based on the adversarial samples; and evaluating the adversarial robustness of the deep learning model based on the class activation maps of the original samples and the adversarial samples before and after training. The method and terminal device of the present invention can mine implicit knowledge in the deep learning model, evaluate adversarial robustness from the perspective of the model itself rather than direct model results, and introduce key position information to enhance the interpretability of the evaluation results.

This invention discloses an unsupervised lesion segmentation method based on large model and pseudo-label learning, belonging to the field of image recognition. First, a "Visual-Language" large model (CLIP) is used to generate high-quality classification pseudo-labels, which are then used to supervise the training of an image classification network. During this process, we further extract the class activation map (CAM) of the classification network and introduce a data augmentation strategy based on dynamic occlusion to further improve the performance of the classification network. Subsequently, we utilize these more accurate class activation maps to generate guidance information for the segmentation large model (SAM), thereby obtaining high-quality segmentation pseudo-labels. Finally, this invention proposes a self-training strategy to train and continuously optimize the segmentation network using these pseudo-labels to achieve higher segmentation accuracy and efficiency.

The invention discloses a high-spatial-resolution remote sensing image solar cell panel identification method based on deep learning. The method comprises the following steps: (1) constructing a high-spatial-resolution RGB satellite image solar cell panel data set containing various complex scenes; (2) carrying out transfer learning twice on the basis of an Inception-v3 network architecture so as to optimize a pre-training weight; (3) realizing visualization and semantic segmentation of the solar cell panel through morphological processing and threshold optimization by using a class activation mapping (CAM) technology; and (4) verifying the recognition effect through evaluation indexes (such as accuracy and F1 score). According to the method, for complex scenes such as roof angles, shadows and sundry shielding, through multi-scale feature extraction of the network and data enhancement of the data set, the precision and robustness of solar cell panel recognition are remarkably improved, and reliable data support is provided for regional solar deployment and social and economic benefit analysis.

This invention discloses a weakly supervised open-pit mining site identification method based on adaptive SAM and automatic knowledge learning. The method includes: S1, collecting images of the study area and extracting three-band images using an input spectral adapter; extracting depth features using the feature extractor of the MobileSAM model; constructing a multi-scale perception and geometric adaptation module and obtaining image features by applying N² dilated convolutional units with different dilation rates to the depth features; S2, a scene classifier using predefined prototype features to guide image features in scene classification; then, a cue generator using class activation mapping to obtain a heatmap for scene classification and selecting a set of high-confidence cue points; and a pseudo-label generator constructing triplet contrast constraints; S3, the mask decoder of the MobileSAM model combining cue information and image features to perform cue decoding and obtain the open-pit mining site identification result. This invention combines a set of high-confidence cue points with image features for cue decoding to obtain efficient and high-precision open-pit mining site identification results.

The application discloses a kind of weak supervision medical image segmentation method based on causality invariance, the application uses two-stage training process, the first stage is label generation, and the picture-level label is generated pixel-level pseudo-label by class activation mapping mechanism, solve the problem that existing algorithm highly depends on a large number of artificial marking pixel-level label.The second stage is segmentation model training, and a segmentation model is trained using the pixel-level pseudo-label output in the first stage, to provide effective segmentation results for the actual clinical application of medical image analysis.Furthermore, a multi-branch feature extraction network is designed in the label generation stage of the application, and a causality invariance conversion strategy is proposed, so that the model can better extract the features of the same target in different states.Meanwhile, the application designs a loss function for different branch networks to ensure the classification accuracy of the model.Finally, the application designs a causality invariance constraint loss between multiple branches to ensure the segmentation accuracy of the model.