6522results about "Biometric pattern recognition" patented technology

An example operation may include at least one of receiving, via a user interface of a device, an activation input from a user to initiate a session, capturing, by a camera of the device, a scan of a body of the user, wherein the capturing comprises recording at least one image and / or at least one video of the user, processing the at least one image and / or video to generate a three- dimensional model of the user comprising measurements and contours of the body, retrieving, from a database, at least one clothing item associated with the user, the at least one clothing item comprising dimensional attributes and texture attributes, rendering, by a graphics processing unit, the at least one clothing item onto the three-dimensional model to generate a visual representation, wherein the rendering simulates draping behavior, movement, and light interaction of the at least one clothing item relative to the three-dimensional model, and displaying, on the user interface, an interactive visualization comprising the visual representation of the three-dimensional model with the at least one clothing item from multiple viewing angles.

The invention relates to the field of breeding industry, and discloses a pig behavior-based pig health condition analysis method and system, and the method comprises the steps: carrying out the comprehensive monitoring of pig behaviors, capturing the gait, feeding mode, excretion behavior and activity range of a pig in real time based on a behavior feature extraction algorithm, and obtaining a behavior state video stream sequence; multi-dimensional time sequence correlation analysis is carried out on the behavior state video stream sequence, and historical behavior data, pig weight changes and physiological parameters are combined; based on the dynamic time sequence feature vector, dynamically identifying a change track of pig behaviors by applying a self-adaptive behavior identification algorithm; performing correlation analysis on the detected abnormal behavior pattern and the potential health risk of the pig, fusing the environmental factors, group behavior data and health history of the pig, and identifying a potential health problem; and based on a risk early warning result, automatically adjusting environmental parameters and feeding management strategies, and providing intervention measure suggestions. The pig health management system has the advantage of improving the efficiency and accuracy of pig health management.

The invention relates to the technical field of security and protection management, in particular to a campus security and protection management system based on deep learning, which improves the accuracy and robustness of identity recognition by acquiring access control card numbers, face images or fingerprint features and generating standardized identity authentication data. And on the basis of a comparison result of the identity authentication data and the campus database, a behavior chain initialization identifier is generated, and accurate identity binding of the school entering personnel is realized. Furthermore, by collecting multi-camera image stream data, pedestrian re-identification and similarity calculation are executed by using a deep feature matching network, and a cross-camera continuous trajectory data set is generated. And matching the behavior track data set with the conventional path template to generate a behavior offset feature vector. And carrying out joint modeling on the behavior offset characteristics and the identity information through a graph neural network model containing an attention mechanism, and outputting a behavior purpose label and a risk grade score. And a graded security response instruction is generated based on the risk score, so that the missing report rate and the false report rate are effectively reduced.

The invention discloses a key place-oriented abnormal behavior intelligent identification and pre-processing system, which is characterized in that a preliminary abnormal event sequence is generated by collecting multi-modal environment data, the preliminary abnormal event sequence and a preset scene knowledge graph are subjected to semantic fusion to form composite abnormal event description information, and then a dynamic processing plan is generated based on large language model reasoning; the central scheduling agent is decomposed into a cooperative control instruction set to drive the video analysis agent, the broadcast grooming agent and the security and disinfection linkage agent to execute cooperative processing operation, situation evolution information is generated in a shared event canvas through environment feedback data, and dynamic optimization and adjustment of a processing strategy are achieved. According to the system, the whole process intelligence of the abnormal event from identification to disposal is realized, the semantic understanding ability of the system to a complex scene and the multi-agent collaborative response efficiency are improved, and the pertinence and the adaptive adjustment ability of a disposal plan are enhanced.

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

The invention relates to the technical field of target tracking, in particular to a photoelectric tracking algorithm and system for self-adaptive target tracking, and the system comprises an image collection module, a cross-spectrum self-adaptive sensing module, an image processing and target recognition module, a self-adaptive tracking algorithm module and a servo control module, and constructs a closed-loop feedback link to achieve the synchronous optimization of parameters in a whole link. The algorithm comprises the steps of image acquisition and adjustment, target detection, parameter optimization, servo driving and feedback. The system can accurately estimate a high-speed turning target, is adaptive to complex environments such as low illumination / haze and the like, completes sheltered target recapture within 0.5 s, has servo compensation precision of + / -0.1 degree, is suitable for airport bird monitoring, port unmanned aerial vehicle tracking and highway vehicle monitoring scenes, and significantly improves tracking stability and practical value.

The invention discloses a human body posture estimation method based on millimeter wave radar point clouds, which comprises the following steps of: processing each frame of input millimeter wave radar sparse point clouds by utilizing a built posture estimation model, and finally predicting and outputting a three-dimensional coordinate sequence of corresponding human body key joint points; wherein the attitude estimation model is composed of a point cloud completion network based on an encoder-decoder architecture, a global-local double-branch network and a feature fusion and prediction mechanism, and the method comprises the following steps: firstly, enhancing the density and integrity of an original sparse point cloud by using the point cloud completion network; the method comprises the following steps: complementing point clouds and original point clouds, respectively processing the complemented point clouds and original point clouds through a global-local double-branch network, extracting global structure features and local detail features, finally fusing the two features through a feature fusion and prediction mechanism, predicting three-dimensional coordinates of human body joint points through a regression layer, and realizing accurate attitude estimation. The method improves the estimation precision and robustness, maintains the privacy protection advantage, and is suitable for complex application scenes.

The invention relates to the technical field of multi-modal posture recognition, and relates to a multi-modal posture recognition system based on Transform fusion, which fully excavates the complementary advantages of visual information in spatial detail representation and inertial information in time sequence dynamic capture through space-time alignment of multi-modal data and deep fusion based on an attention mechanism. The accuracy and the stability of attitude estimation under the conditions of visual occlusion, rapid movement and complex illumination are obviously improved; posture optimization is carried out by introducing physical constraints such as bone length constancy and joint movement range limitation, and time domain smoothing and contact state correction are applied, so that the precision of the generated three-dimensional posture sequence is ensured, a solid technical foundation is laid for improving the naturalness, safety and intelligent level of man-machine interaction, and the method is suitable for popularization and application. And meanwhile, reliable application and deep development of the intelligent robot in key fields of service robots, virtual reality, remote cooperation and the like are powerfully promoted.

The invention discloses a cable tunnel fire risk feature identification system and identification method, and belongs to the technical field of cable tunnel safety monitoring. The whole stage of a fire is covered through multi-class cooperative detection, the target identification precision and scene adaptability are greatly improved, and the safety of the cable tunnel is improved. According to the whole system, a flame and smoke detection unit and a flame / smoke special detection unit are innovatively added to a detection engine module, an original heat source detection unit and an original human body detection unit are combined, a heat source-flame-smoke-human body four-category collaborative detection framework is formed, high-precision recognition is achieved on the basis of a YOLO model framework, and the detection efficiency is improved. Compared with the problems that a traditional system is high in single target detection omission ratio and cannot cover the whole stage of smoldering-initial open fire-violent combustion of a fire, the system has the advantages that the recognition rate of early flame and weak smoke is increased, the false alarm rate is greatly reduced, meanwhile, a heat source of operation and maintenance personnel is prevented from being misjudged as a fire hazard through human body detection, and the safety of the fire hazard is improved. And cable monitoring and personnel safety protection are both considered.

The invention discloses a poultry behavior abnormity real-time monitoring system based on multi-modal image fusion, particularly relates to the technical field of intelligent breeding behavior recognition, and is used for solving the problem of poor behavior monitoring accuracy under feather shielding. The method comprises the following steps: firstly, through combined perception of a visible light image and an infrared image, extracting a claw track interruption point and an anus temperature gradient direction, and realizing analysis of a motion state of a sheltered area; then, in combination with the heat conduction delay characteristic and the group movement direction, the flexion and extension angle of the covered leg joint is inverted, and a complete gait sequence is generated; thirdly, multi-source features such as gaits, temperature differences and body postures are fused, and a dynamic deviation model of the individuals relative to the mass center of the group is constructed; and finally, generating a stress behavior threshold curve according to the ground temperature and the ammonia gas concentration, outputting an abnormal behavior type and confidence, and realizing intelligent distinguishing of mechanical obstacles and adaptive behaviors.

The invention relates to the technical field of computer vision and cross-visual-angle pedestrian recognition, in particular to a cross-visual-angle pedestrian re-recognition method based on visual language prompt learning, and the method comprises the steps: obtaining a target image; the target image is input into a preset pedestrian recognition model, a pedestrian re-recognition result is output, the pedestrian recognition model is obtained by training a visual language pre-training model CLIP through a prompt learning mechanism and a two-stage training strategy, the prompt learning mechanism is used for modeling visual angle deviation, and the two-stage training strategy is used for training visual angle deviation. The double-stage training strategy is used for realizing cross-modal semantic alignment. According to the invention, the accuracy and robustness of cross-view identification can be significantly improved.

The invention discloses a human body infrared image small target detection method based on an improved FGLCM feature, and relates to the technical field of image processing and target detection, and the method comprises the following steps: building a boundary singularity auditing baseline under a unified time baseline, carrying out the multi-scale energy mapping of a curved surface fitting residual error of a human body infrared image, and generating a traction residual error distribution diagram; and constructing a reflection pseudo peak discriminator based on the traction residual distribution diagram, and extracting a pseudo peak kernel position by combining polarization sensitivity estimation and view angle transformation consistency constraint. According to the method, a closed-loop self-adaptive regulation and control mechanism is constructed through residual distribution auditing, pseudo peak identification, gradient registration, differential entropy enhancement and time sequence threshold adjustment, fitting abnormity and false highlight spots in the infrared image are inhibited, the accuracy and stability of small target detection are improved, and the method is suitable for a complex photo-thermal environment.

The invention provides a sign language translation method and system based on a pre-training diffusion large language model, and belongs to the field of sign language video translation. The method comprises the following steps: preprocessing a video containing sign language actions to obtain a sign language video frame sequence, inputting the sign language video frame sequence into a visual feature extraction network to extract features, and fusing to obtain a time sequence visual fusion feature sequence; giving a text cue word of a sign language translation task, constructing an initial mask sequence for a target translation position, taking the text cue word, the time sequence visual fusion feature sequence and the initial mask sequence as guide conditions, injecting the guide conditions into a diffusion language model, iteratively denoising and predicting lexical elements of a masked position in combination with a diffusion mask mechanism, and obtaining the sign language translation task. A natural language translation sequence is obtained, and sign language translation is completed; wherein when the diffusion language model is trained, through an internal feature alignment mechanism, the guiding effect of guiding conditions on text generation is optimized, so that the accuracy, coherence and robustness of long text translation are improved, and the actual requirements of a barrier-free public service scene are better met.

Disclosed is a image processing apparatus and a method for controlling the image processing apparatus. The image processing apparatus according to an embodiment of the present disclosure may identify an object from an acquired image, determine whether the object is hidden by another object by using an aspect ratio of a bounding box of the detected object, and based on the object being hidden, estimate an entire length of the object based on coordinate information of the bounding box. Accordingly, the size information of the hidden object may be efficiently identified while a large amount of database is applied or resources of the apparatus is minimized. The present disclosure may be in connection with a surveillance camera, an automotive driving vehicle, an artificial intelligence module of at least one of a user terminal or a server, a robot, an augmented reality (AR) device, a virtual reality (VR) device, a device related to a 5G service, and the like.

The invention discloses a wildlife detection method fusing an unmanned aerial vehicle thermal infrared image and a visible light image, and belongs to the field of small target wildlife identification, and the method comprises the following steps: S1, obtaining a preprocessed TIR-RGB image pair set; s2, an FDM-YOLO double-source target detection model improved based on YOLOv81 is constructed, and the improved FDM-YOLO double-source target detection model is trained based on the preprocessed TIR-RGB image pair set obtained in the step S1; and S3, inputting an image acquired in real time into the improved FDM-YOLO double-source target detection model trained in the step S2, and outputting a wild animal detection result. By adopting the wild animal detection method fusing the thermal infrared image and the visible light image of the unmanned aerial vehicle, high-precision, real-time and robust detection of a small target of a wild animal in a complex field environment is realized by improving the FDM-YOLO model.

The invention discloses an aquaculture comprehensive guarantee method and system based on multi-modal data acquisition, and relates to the technical field of intelligent aquaculture, and the method comprises the steps: collecting aquaculture multi-modal data and an aquaculture image of an aquaculture region, carrying out the aquaculture feature capture of the aquaculture image, and outputting a visual feature group; inputting the visual feature group into a MobileNet lightweight model, and outputting a health state label and a confidence score; based on the health state label and the confidence score, outputting disease type data through a residual network structure optimized by transfer learning; constructing a graph neural network according to the disease type data, and generating a dynamic regulation and control scheme; and the control center executes the dynamic regulation and control scheme, carries out regulation and control effect detection and target comparison on the aquaculture area of the dynamic regulation and control scheme, judges whether the regulation and control effect is effective or not, and generates a feedback regulation instruction. According to the invention, through multi-modal data fusion and intelligent decision closed loop, dynamic and accurate guarantee of aquatic product health management is realized.

The invention relates to the technical field of biomedical engineering and computer vision, in particular to a non-contact physiological signal extraction method and system based on frequency self-adaption and illumination noise perception.The method comprises the following steps of multi-mode video stream collection and spatio-temporal data preprocessing, illumination-noise perception mask generation and feature filtering, multi-mode video stream collection and spatio-temporal data preprocessing, illumination-noise perception mask generation and feature filtering, and non-contact physiological signal extraction. Frequency adaptive gating and frequency domain feature enhancement, depth time attention feature re-calibration, physiological signal regression and closed loop optimization; the method has the beneficial effects that a lightweight end-to-end deep learning network architecture is constructed by systematically fusing three core modules of illumination-noise perception mask, frequency adaptive gating and depth time attention, and the defects that a traditional physical model depends on artificial prior and is poor in anti-interference performance and high in reliability are overcome. And the one-sidedness caused by high calculation complexity and difficulty in distinguishing the signal and noise of the existing deep learning model is avoided, and the weak physiological signal can be recovered from the face video more accurately and robustly.

The invention discloses a bird identification method based on sound-image multi-modal fusion. The bird identification method comprises the following steps: S1, carrying out standardized frame-level preprocessing on bird audio signals; s2, acoustic features are extracted and enhanced, and an acoustic high-level feature vector which highlights birdsong discrimination information and suppresses environmental noise is obtained; s3, visual image standardization preprocessing; s4, performing visual feature extraction and multi-scale fusion to obtain a visual high-level feature vector which enhances correspondence to the bird key form area and inhibits background interference; s5, performing dynamic weighted fusion on the decision-making layer to obtain a bird existence probability; and S6, comparing the bird existence probability with a preset threshold value of the corresponding bird, and judging whether the bird exists or not and the type of the existing bird. Through cross-modal feature enhancement and adaptive fusion, the precision, robustness and real-time performance of bird recognition in a complex orchard environment are significantly improved, and a core technical support is provided for green intelligent bird repelling.

The invention discloses a creation content generation system based on image recognition and large language model fusion, and particularly relates to the technical field of creation content generation, and the system firstly completes the fact extraction and brand anchor point construction of an input image in a unified coordinate and scale system, and forms a structured fact package in one-to-one correspondence with an original image; then, performing protagonality scoring and ambiguity gating on the figure instance, and outputting an explainable and calibratable protagonality judgment result; on this basis, the condition controlled generation and template selection module converts the fact constraint into a controlled text packet and a format instruction packet, and keeps explicit mapping with a fact packet; the system further executes cross-modal consistency and compliance verification based on image facts, and machine-readable verification and minimum cost correction are carried out on text and graph entities, geometrical relationships and brand elements; and finally, solidifying the key intermediate quantity, the parameters and the judgment basis into an evidence chain through a chain type index, and introducing online adaptive learning in a compliance boundary to realize mild updating and rollback release.

The invention discloses a millimeter-wave radar sparse point cloud-based human body posture recognition method, which belongs to the technical field of human body posture recognition, and comprises the following steps of: acquiring three-dimensional point cloud data of human body actions through a millimeter-wave radar, and preprocessing the three-dimensional point cloud data; and training a lightweight neural network model by using the preprocessed point cloud data, wherein the model comprises an edge convolution module and a grouping sparse Transform encoder module. The edge convolution module extracts spatial geometric features and detects dynamics, static postures are directly classified, dynamic postures capture a time sequence dependency relationship through a grouping sparse Transform module, attention calculation is only carried out on frames with feature changes exceeding a threshold value, and mean pooling aggregation is carried out on other frames. And finally, classifying the human body postures based on the spatial geometric features and the time sequence dependency relationship to obtain a classification result. The device is simple in structure, accurate in recognition and suitable for efficient deployment of edge equipment.

The present application provides a cardiac fibrosis diagnosis model based on multi-task attentional feature fusion. The cardiac fibrosis diagnosis model is established by the following steps: S01: image collection and labeling: obtaining cardiac magnetic resonance (MR) images as sample data, and performing manual labeling to obtain heart labels corresponding to the MR images; S02: image preprocessing, including normalization processing, data enhancement, and data clipping; S03: model establishment, including establishment of an image recovery network and establishment of an image segmentation and classification network, and executing an image recovery task; S04: model pre-training: training the image recovery network such that the encoder of the image recovery network fully learns the feature of the cardiac fibrosis image; and S05: model training. An objective of the present application is to improve the segmentation precision and diagnosis accuracy of a network model for a cardiac fibrosis image.

An XR system is provided. This system captures images including images of a first hand of a user and a second hand of the user using one or more cameras. The XR system generates cropped images using the images, each cropped image including a surface of the first hand. The XR system detects a hand touch of the surface of the hand by a digit of the second hand using the cropped images. The hand touch is used as an input into an XR user interface of the XR system. The surface of the hand can be palmar surface or a hand dorsal surface.

The invention discloses an intelligent supervision method based on a BIM technology, and relates to the technical field of constructional engineering, and the method comprises the following steps: obtaining a BIM model corresponding to the constructional engineering, constructing a digital twinning mapping relation, and carrying out the construction of a digital twinning mapping relation based on a project type, construction complexity and historical supervision data; initializing a data entry time limit through an algorithm, and changing a synchronous threshold value and a quality parameter benchmark; and collecting real-time data, wherein the real-time data comprises process progress data, material entering data, design change information and environmental safety data of the construction site. According to the intelligent supervision method based on the BIM technology, digital and intelligent transformation of supervision work is realized through deep fusion of the BIM technology and multi-source data acquisition, the matching degree of data and an actual construction scene is improved, the accuracy of anomaly recognition is enhanced through a dynamic weight distribution strategy, and the accuracy of anomaly recognition is improved. The problem of dependence on fixed threshold judgment is effectively solved, and the manual intervention cost is reduced.

The invention relates to a fitness action error correction method and device for bone key point detection, equipment and a medium. The method comprises the following steps: acquiring video data in a multi-person fitness scene, and performing skeleton key point detection on each frame of image in the video data to obtain a multi-person skeleton key point sequence; performing spatial clustering processing on the multi-person skeleton key point sequence to obtain a plurality of individual skeleton key point sequences; according to the skeleton key point sequence of each individual, generating an action track of each individual; based on a preset standard action sequence, identifying an action matching result corresponding to each action track; according to the action matching result, the action deviation value of each individual is calculated, and action error correction feedback information is generated based on the action deviation values; the error correction feedback information is used for indicating each individual to adjust the fitness action. The method can adapt to complex scenes, solves the problem that key points of scenes of multiple persons are confused, and provides effective support for fitness guidance and the like.

The invention provides a personnel trajectory tracking method and system, and belongs to the technical field of personnel trajectory tracking, and the method comprises the steps: obtaining first data and second data; processing the first data to obtain a personnel bounding box set in the first data; processing the second data to generate a second coordinate system; obtaining a personnel point cloud clustering set in the second coordinate system; based on the personnel bounding box set and the personnel point cloud clustering set, determining a target personnel set through joint confidence calculation; calculating the current three-dimensional position of each target person in the second coordinate system as a current observation node; constructing a space-time association graph based on the current observation node and the historical trajectory node of the target person; and tracking a continuous motion track of the target person based on the space-time association diagram. According to the invention, cross-modal personnel target efficient alignment and continuous trajectory tracking can be realized.

The invention discloses a power worker behavior detection method and related equipment. The method comprises the following steps: acquiring first data in a scene where a worker is located through a camera in a working site; the first data comprises continuous frame images or a video sequence lasting for preset time; analyzing the first data to obtain a current first identification result of the operator; the first identification result is used for representing whether the operator lacks of wearing the protective equipment and / or carries illegal articles; determining the current posture of the operator from the first data; and fusing the first recognition result and the current posture to obtain the operation risk level of the operator. The method can be widely applied to the technical field of artificial intelligence.

The invention discloses an occluded pedestrian re-identification method based on cross-layer frequency domain enhancement and multi-view fusion, and the method comprises the steps: employing ResNet-50 as a backbone network, inputting a plurality of pedestrian images of the same identity according to groups, obtaining multiple layers, carrying out the frequency separation and frequency enhancement, forming a significant mask which is more sensitive to the occlusion and background, and carrying out the recognition of the pedestrian images. And meanwhile, the lay1 obtains global features with the same scale as the lay4 through GCSA, and weighted fusion is carried out on the global features and the lay4 features according to the occlusion score, so that the occluded area is completed. Group-level representation with multi-view attention fusion, output information complementation and noise suppression is adopted. In the training stage, the identity classification loss of each branch and the cross-branch consistency loss are jointly optimized, and gradient cutting is matched to improve the stability. According to the method, on the premise that a ResNet-50 backbone structure is not changed, cross-layer frequency domain prior is used for accurate shielding positioning, global-local adaptive fusion based on shielding scores and multi-view weighting are combined, and the pedestrian re-recognition precision and robustness in a shielding scene are remarkably improved.

A marine zooplankter identification method based on in-situ images and deep learning is used for processing a multi-stage series deep learning architecture constructed for all in-situ images, and comprises the following steps: quickly positioning and framing zooplankter individuals under a complex background based on a YOLOv8 skeleton construction model; performing pixel-level fine segmentation on individuals in the in-situ image by using a U-Net model, and extracting an accurate contour to obtain a high-quality individual image; and finally, inputting the image into a PlanktonNet network innovatively designed based on a ViT architecture, and through adaptive small-size slice embedding and introduction of a convolution word embedding layer, improving small-scale target feature extraction capability, and realizing high-precision and fine-grained classification of genera and species. The method has the advantages that target detection, semantic segmentation and recognition tasks are organically fused, the problems of low recognition efficiency and low automation degree in the prior art are effectively solved, and marine zooplankton can be quickly and accurately recognized from in-situ images on a large scale.

The invention relates to the technical field of artificial intelligence, in particular to a video GIS intelligent analysis method based on deep learning, and the method comprises the steps: detecting a target in a video in real time through a predefined target detection model, and generating a space-time mark which comprises a bounding box and category information; thirdly, associating target tracks of different cameras by using a dynamic graph model and a dynamic graph neural network to form a space-time ID and a track chain; then, combining a visual inertial odometer and a geographic information system to calibrate a homography matrix, and mapping the trajectory chain to a geographic coordinate system to obtain space-time trajectory data; then, constructing a neural network model and a trajectory generation model, and respectively predicting crowd density distribution and pedestrian motion trajectories; and finally, carrying out real-time alarm according to the track and the multi-level geo-fencing rule. According to the invention, intelligent analysis of video data is realized, target tracking and alarm capabilities are improved, and the method is widely applicable to the fields of crowd management and safety monitoring.

The invention relates to the technical field of pedestrian re-identification, and particularly provides an end-to-end cross-modal pedestrian re-identification method based on multi-domain feature alignment. The method comprises the following steps: splitting original text description into two sub-descriptions of identity description and clothing description through a text description separation module, and providing structured semantic input for cross-modal feature alignment; according to the two sub-descriptions of the identity and the clothing, the identity-clothing bidirectional decoupling alignment module utilizes an attention mechanism and a gating weighting strategy to realize cross-modal feature alignment; based on cross-modal feature alignment, introducing a Mama state space model SSM into a cross-modal pedestrian re-identification ReID task, and fusing image and text features; according to the features of the fused image and the text, a multi-target robust optimization module is designed for optimization, and a final retrieval result is output, the precision of fine-grained semantic alignment is improved, effective context collaboration is achieved, and balance between discrimination and robustness is achieved.