Object recognition method and device, computer equipment and storage medium

Abstract

The invention relates to the technical field of artificial intelligence and the field of medical health, and discloses an object recognition method and device, computer equipment and a storage medium, and the method comprises the steps: determining a to-be-recognized target object, and collecting an image containing the target object in real time through image collection equipment; preprocessing the collected image to obtain a target image; extracting all objects in the target image by adopting an image segmentation algorithm; performing feature recognition on each object by using a visual large model, and converting the recognized features into feature descriptions; based on the feature description of each object, generating an identification result of the target object through a large language model; therefore, the problem that in the prior art, an existing object recognition method is insufficient in accuracy, robustness, expandability and the like can be solved.

Description

Technical Field

[0001] The present invention relates to the fields of artificial intelligence technology and medical health, and in particular to an object recognition method, device, computer equipment and computer-readable storage medium. Background Art

[0002] At present, in the existing technology, object recognition systems mainly rely on traditional image processing algorithms and machine learning models. Although these methods can achieve certain recognition effects under certain conditions, they have significant limitations in many aspects:

[0003] 1. Environmental adaptability: Traditional methods are very sensitive to environmental changes, such as lighting conditions, background complexity, object occlusion and posture changes, which may lead to a significant decrease in object recognition performance;

[0004] 2. Feature generalization ability: Many existing algorithms perform poorly in feature generalization and have difficulty adapting to new or unseen objects, limiting their application in diverse scenarios;

[0005] 3. Scalability and maintainability: With the increase of new types of objects, existing object recognition systems often need to re-collect data and train models. This process is time-consuming and costly, affecting the scalability and maintainability of the system.

[0006] 4. Accuracy and robustness: In complex or dynamically changing environments, traditional object recognition systems are often insufficient in accuracy and robustness, and are unable to meet current application scenarios with high precision requirements.

[0007] In summary, in the prior art, current object recognition methods have deficiencies in accuracy, robustness, and scalability. These deficiencies are particularly prominent in fields such as medical health that require extremely high precision.

[0008] Therefore, the existing technology still needs to be improved and developed. Summary of the invention

[0009] In view of the above-mentioned deficiencies in the prior art, the object of the present invention is to provide an object recognition method, apparatus, computer device and computer-readable storage medium, aiming to solve the problems in the prior art that the current object recognition method has deficiencies in accuracy, robustness and scalability.

[0010] In order to achieve the above object, the present invention adopts the following technical solutions:

[0011] In a first aspect, the present invention provides an object recognition method, which comprises:

[0012] Determine a target object to be identified, and use an image acquisition device to acquire an image containing the target object in real time;

[0013] Preprocessing the collected image to obtain a target image;

[0014] Using an image segmentation algorithm to extract all objects in the target image;

[0015] Using the visual macromodel to perform feature recognition on each of the objects, and converting the recognized features into feature descriptions;

[0016] Based on the feature description of each of the objects, a recognition result of the target object is generated through a large language model.

[0017] In a second aspect, the present invention provides an object recognition device, comprising:

[0018] An acquisition module, used to determine a target object to be identified, and to acquire an image containing the target object in real time using an image acquisition device;

[0019] A preprocessing module, used for preprocessing the collected image to obtain a target image;

[0020] An extraction module, used for extracting all objects in the target image by using an image segmentation algorithm;

[0021] A conversion module, used for performing feature recognition on each of the objects using a large visual model, and converting the recognized features into feature descriptions;

[0022] A result generation module is used to generate a recognition result of the target object through a large language model based on the feature description of each object.

[0023] In a third aspect, the present invention provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the object recognition method as described above when executing the computer program.

[0024] In a fourth aspect, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, wherein the computer program implements the object recognition method as described above when executed by a processor.

[0025] Compared with the prior art, the present invention provides an object recognition method, apparatus, computer equipment and computer-readable storage medium, wherein a target object to be recognized is determined, and an image containing the target object is captured in real time using an image capture device; the captured image is preprocessed to obtain a target image; an image segmentation algorithm is used to extract all objects in the target image; a large visual model is used to perform feature recognition on each of the objects, and the recognized features are converted into feature descriptions; based on the feature descriptions of each of the objects, a recognition result of the target object is generated through a large language model; thereby, the present invention can solve the problem in the prior art that the current object recognition method has deficiencies in accuracy, robustness and scalability. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0027] Figure 1 A schematic diagram of an application environment of an object recognition method provided by an embodiment of the present invention.

[0028] Figure 2 A flowchart of an object recognition method provided by an embodiment of the present invention.

[0029] Figure 3 A schematic diagram of a program module of an object recognition device provided by an embodiment of the present invention.

[0030] Figure 4 A schematic diagram of the structure of a computer device provided by an embodiment of the present invention.

[0031] Figure 5 Another structural schematic diagram of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0032] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0033] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0034] It should also be understood that the term "and / or" used in the present description and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0035] As used in the present specification and the appended claims, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0036] In addition, in the description of the present specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0037] References to "one embodiment" or "some embodiments" etc. described in the present specification mean that one or more embodiments of the present invention include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0038] It should be understood that the order of execution of the steps in the following embodiments does not imply a precedence of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0039] In order to illustrate the technical solution of the present invention, specific embodiments are provided below for illustration.

[0040] An object recognition method provided by an embodiment of the present invention can be applied in Figure 1In the application environment shown, the client and the server communicate through the network. The client includes but is not limited to PDAs, desktop computers, laptops, ultra-mobile personal computers (UMPCs), netbooks, cloud computing devices, personal digital assistants (PDAs) and other computer devices. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0041] See also Figure 2 An embodiment of the present invention provides an object recognition method, wherein the method comprises the following steps:

[0042] S100, determining a target object to be identified, and using an image acquisition device to acquire an image containing the target object in real time;

[0043] S200, preprocessing the collected image to obtain a target image;

[0044] S300, extracting all objects in the target image using an image segmentation algorithm;

[0045] S400, using the visual large model to perform feature recognition on each of the objects, and converting the recognized features into feature descriptions;

[0046] S500: Based on the feature description of each of the objects, generate a recognition result of the target object through a large language model.

[0047] In specific implementation, this embodiment effectively improves the accuracy, robustness and scalability of object recognition by integrating advanced image processing technology and artificial intelligence models, and solves some related problems in the prior art. Specifically:

[0048] 1. Improved accuracy: By using the image segmentation algorithm (step S300), the target object can be accurately extracted from the complex background, reducing the impact of background noise and interference on the recognition process; using the visual large model (step S400) for feature recognition, these models usually have deep convolutional networks, which can capture more complex and abstract features, thereby improving the accuracy of recognition;

[0049] 2. Enhanced robustness: The preprocessing step (step S200) can adjust the contrast and brightness of the image and perform denoising and other operations, so that the subsequent recognition process has better adaptability to adverse conditions such as lighting changes and noise; large language models (step S500) are used to generate recognition results. These models have been pre-trained on large-scale data sets and have strong generalization capabilities, and can maintain stable recognition performance in different objects and scenes;

[0050] 3. Enhanced scalability: The modular design of the method in this embodiment allows each step to be replaced or upgraded as needed. For example, more advanced image segmentation algorithms or more powerful visual and language models can be used to adapt to changing recognition needs. The generation of recognition results using large language models can be easily expanded to new object categories and scenes because these models already have rich semantic understanding capabilities.

[0051] To sum up, the object recognition method of this embodiment not only improves the accuracy and robustness of recognition, but also enhances the scalability and adaptability of the object recognition method by combining image preprocessing, advanced image segmentation technology, deep learning visual model and large language model, thereby solving the shortcomings of the existing technology in terms of technical effects.

[0052] Furthermore, in one embodiment, the object recognition method, wherein the step S100 of determining the target object to be recognized and using an image acquisition device to acquire an image containing the target object in real time, specifically comprises the steps of:

[0053] Determine the target object to be identified according to the task objective;

[0054] Determine the image acquisition device based on the task environment;

[0055] The image acquisition device is used to acquire images containing the target object in real time.

[0056] In specific implementation, the technical effects that can be achieved by this embodiment include ensuring the efficiency and adaptability of image acquisition, and improving the accuracy and reliability of subsequent object recognition; first, determining the target object to be identified through clear task objectives helps to optimize the image acquisition and processing strategy, and ensure that the acquired image contains the key information required to complete the specific task; second, selecting a suitable image acquisition device according to the task environment can ensure that high-quality images are obtained under specific lighting, distance and angle conditions, thereby adapting to different working environments; finally, using the selected image acquisition device to capture images containing target objects in real time, providing a continuous and real-time data stream for object recognition, which is particularly important for dynamic monitoring and rapid response application scenarios; this embodiment combines these steps to improve the recognition efficiency of target objects and ensure high-performance object recognition capabilities in changing actual application environments.

[0057] The specific implementation process of the steps in this embodiment is roughly as follows:

[0058] 1. Requirements analysis and goal definition:

[0059] Task target analysis: First, analyze the specific requirements of the task to understand the type of target object to be identified (such as pedestrians, vehicles, products, etc.) and the purpose of identification (such as counting, classification, positioning, etc.);

[0060] Target object characteristic analysis: Study the physical and visual characteristics of the target object, such as size, color, shape, texture, etc., to provide a basis for subsequent image acquisition and processing;

[0061] 2. Environmental assessment and equipment selection:

[0062] Mission environment assessment: examine the environmental conditions for mission execution, including factors such as lighting, distance, angle, dynamic changes, etc.

[0063] Equipment selection: Based on the results of the environmental assessment and the characteristics of the target object, select the appropriate image acquisition equipment; for example, an infrared camera may be required in low-light environments, and a high-pixel camera may be required when high-resolution details are required;

[0064] 3. Equipment configuration and testing:

[0065] Equipment adjustment: Adjust the selected image acquisition equipment to the appropriate position to ensure that it can cover the area where the target object may appear;

[0066] Parameter adjustment: adjust the parameters of the device, such as focus, exposure, white balance, etc., to adapt to environmental conditions and improve image quality;

[0067] Pre-test: Conduct preliminary image acquisition tests to check whether the image quality meets the requirements, and further adjust the equipment configuration based on the test results;

[0068] 4. Real-time image acquisition:

[0069] Start acquisition: Start the image acquisition device to perform real-time image acquisition;

[0070] Image monitoring: real-time monitoring of collected images to ensure image quality and continuity;

[0071] Data management: store or transmit the collected image data for subsequent image processing and analysis;

[0072] 5. Quality control and feedback adjustment:

[0073] Image quality assessment: Regularly assess the quality of acquired images, including clarity, contrast, noise level, etc.

[0074] Equipment status monitoring: monitor the operating status of image acquisition equipment to promptly detect and resolve possible faults;

[0075] Parameter optimization: Dynamically adjust the parameters of the acquisition equipment based on the results of quality assessment and equipment status monitoring to maintain the optimal performance of image acquisition;

[0076] Through the above process, this embodiment can ensure that the image acquisition process can efficiently and accurately capture images containing target objects, providing high-quality data support for subsequent object recognition and analysis.

[0077] Furthermore, in one embodiment, the object recognition method, wherein the step S200 of preprocessing the acquired image to obtain the target image, specifically comprises the steps of:

[0078] After performing image denoising, image enhancement and image conversion on the collected image, a preliminary image is obtained;

[0079] The preliminary image is subjected to quality inspection, and when the inspection result meets the preset image quality requirement, the target image is obtained.

[0080] In specific implementation, the technical effect that can be achieved by this embodiment is to improve the overall quality of the image through a series of image preprocessing operations, ensuring the accuracy and reliability of the subsequent analysis and recognition process. First, image denoising helps to eliminate random noise that may be introduced during the image acquisition process, making the image clearer; image enhancement makes the features of the target object more prominent and easy to identify by adjusting parameters such as contrast and brightness; image conversion may include color space conversion, size scaling, etc. to adapt to specific processing procedures or meet specific analysis requirements. Then, through the quality detection step, the preliminary images are evaluated to ensure that they meet the preset quality standards, such as clarity, contrast, and integrity, so as to screen out images suitable for the next step of processing. In this way, only images that meet the quality requirements will be identified as target images for subsequent object recognition or other visual analysis tasks, thereby effectively improving the robustness and accuracy of the recognition system.

[0081] The specific implementation process of the steps in this embodiment is roughly as follows:

[0082] 1. Image preprocessing:

[0083] Denoising: Apply denoising algorithms, such as Gaussian filtering, median filtering, or bilateral filtering, to reduce random noise in the image and improve image quality;

[0084] Enhancement: By adjusting brightness, contrast or using histogram equalization, the visual effect of the image is enhanced to make the features of the target object more obvious;

[0085] Transformation: Perform necessary image transformations such as color space conversion (e.g., from RGB to grayscale or HSV), rotation, scaling, or cropping to meet analysis needs;

[0086] 2. Preliminary image generation:

[0087] Synthesis processing: combine the results of denoising, enhancement and conversion to form a preliminary image;

[0088] Storage: Save the preliminary images in appropriate storage media for subsequent quality inspection and further processing;

[0089] 3. Image quality detection:

[0090] Define standards: pre-define image quality standards and parameters such as resolution, contrast, sharpness, etc. according to application requirements;

[0091] Automatic detection: using image quality assessment algorithms, such as those based on blurriness, noise level, or structural similarity (SSIM), to automatically detect whether the preliminary image meets the preset image quality requirements;

[0092] 4. Target image determination:

[0093] Qualification judgment: If the detection result of the preliminary image meets the preset image quality requirements, it will be determined as the target image;

[0094] Feedback mechanism: If the detection results do not meet the requirements, adjust the preprocessing parameters or re-acquire the image based on the feedback;

[0095] 5. Preparation for subsequent applications:

[0096] Labeling: labeling or classifying target images to facilitate subsequent object recognition or other visual analysis tasks;

[0097] Transmission: Transmit the target image to a processing server or cloud platform for further analysis and recognition;

[0098] Through the above process, this embodiment can ensure that only high-quality target images are used for subsequent analysis and recognition, thereby improving the performance and reliability of the entire visual system.

[0099] Furthermore, in one embodiment, the object recognition method, wherein the step S300, extracting all objects in the target image using an image segmentation algorithm, specifically comprises the steps of:

[0100] Determining an image segmentation algorithm according to image content features of the target image;

[0101] The image segmentation algorithm is used to extract all objects in the target image, and a corresponding unique identifier is assigned to each of the extracted objects.

[0102] In specific implementation, the technical effect that can be achieved by this embodiment is to effectively separate each object from the target image by accurately selecting an image segmentation algorithm that is suitable for the image content features of the target image, and assign a unique identifier to each object, thereby laying a solid foundation for subsequent object recognition, tracking and analysis. This embodiment improves the accuracy and efficiency of segmentation, ensures that a single object can be accurately identified and processed even in complex scenes, and at the same time, through the assignment of unique identifiers, enhances the ability to further operate and manage objects, whether in the fields of automated monitoring, robot navigation or content recognition, significantly improving the performance of the object recognition system and the flexibility of application.

[0103] The specific implementation process of the steps in this embodiment is roughly as follows:

[0104] 1. Image content feature analysis:

[0105] Feature extraction: First, analyze the target image and extract key content features that affect the image segmentation effect, such as color, texture, shape, and size;

[0106] Algorithm selection: According to the extracted content features, select the most suitable image segmentation algorithm, such as threshold-based segmentation, edge detection, region growing, deep learning segmentation network, etc.;

[0107] 2. Application of image segmentation algorithm:

[0108] Algorithm configuration: configure the selected image segmentation algorithm and set necessary parameters, such as seed points, thresholds or network weights;

[0109] Segmentation execution: Apply the configured algorithm to the target image, perform the segmentation process, and separate the objects in the target image;

[0110] 3. Object extraction and identification:

[0111] Object recognition: Identify all independent objects from the segmentation results;

[0112] Identity assignment: assign a unique identifier to each extracted object, which can be a number, a label, or a more complex data structure;

[0113] 4. Result verification and optimization:

[0114] Verify the segmentation effect: Check whether the segmentation results meet the requirements and confirm whether all objects are correctly segmented and identified;

[0115] Algorithm adjustment: If the segmentation effect is not good, adjust the algorithm parameters or change the image segmentation algorithm based on the feedback;

[0116] 5. Data recording and output:

[0117] Record segmentation information: record the segmentation information and corresponding unique identifier of each object for subsequent use;

[0118] Output results: Output the segmented objects and their unique identifiers into an appropriate format for subsequent processing;

[0119] Through the above process, this embodiment can ensure the accuracy and efficiency of the image segmentation process and provide a clear foundation for subsequent object processing work.

[0120] Furthermore, in one embodiment, the object recognition method, wherein the step S400, using the visual macro model to perform feature recognition on each of the objects and converting the recognized features into feature descriptions, specifically comprises the steps of:

[0121] Pre-build visual models;

[0122] Performing feature recognition on each of the objects through the visual large model, and converting the recognized features of each of the objects into feature text;

[0123] The feature text is sorted to generate a feature description.

[0124] In specific implementation, the technical effect that can be achieved by this embodiment is to use the pre-built visual big model to perform deep feature recognition on objects and effectively convert these features into understandable text descriptions. This process first extracts the key visual features of each object through the visual big model, and then enhances the interpretability and readability of the features by converting these feature information into text form. Finally, by organizing and formatting the feature text, a clear and structured feature description is generated, which not only improves the accuracy of object recognition, but also facilitates subsequent data analysis, information retrieval and report generation, and enhances the interactive capabilities and user experience of the entire object recognition system.

[0125] The specific implementation process of the steps in this embodiment is roughly as follows:

[0126] 1. Build a large visual model:

[0127] Model selection: Choose a suitable deep learning architecture, such as convolutional neural network (CNN), for visual feature extraction;

[0128] Training data preparation: Collect and annotate a large amount of image data to train the model to recognize the features of various objects;

[0129] Model training: Use the labeled data set to train the model until the model achieves a satisfactory recognition accuracy;

[0130] 2. Feature recognition:

[0131] Image input: Input the image of the object to be identified into the trained visual model;

[0132] Feature extraction: The visual big model analyzes the object image and extracts the features of the object in the image, such as shape, color, texture, etc.

[0133] Feature conversion: converting the extracted feature vectors into interpretable text labels or descriptions;

[0134] 3. Feature text generation:

[0135] Label mapping: Map feature vectors to predefined labels or vocabulary to generate feature text;

[0136] Natural language description: Use natural language processing technology to convert feature text into more fluent and descriptive natural language text;

[0137] 4. Feature text sorting:

[0138] Data cleaning: remove irrelevant or redundant content in feature text to ensure the accuracy and conciseness of feature description;

[0139] Structuring: Organize feature text into a structured format, such as feature lists or attribute-value pairs, to facilitate understanding and use;

[0140] 5. Generate feature description:

[0141] Description writing: Write a detailed and accurate description of the object's features based on the sorted feature text;

[0142] Review and optimization: Review the generated feature description to ensure that it conforms to the features of the actual object and make necessary optimizations;

[0143] Through the above process, this embodiment can build a visual system that can recognize object features and generate easy-to-understand feature descriptions, thereby providing support for various application scenarios, such as object recognition, image retrieval, automated monitoring, and intelligent interaction.

[0144] Furthermore, in one embodiment, the object recognition method, wherein the step S500, based on the feature description of each object, generates the recognition result of the target object through a large language model, specifically comprises the steps of:

[0145] Pre-build large language models;

[0146] Inputting the feature description of each of the objects into the large language model to generate a recognition result of the target object;

[0147] Based on the recognition result, position information of the target object in the target image is determined.

[0148] Furthermore, the object recognition method, wherein after determining the position information of the target object in the target image based on the recognition result, further specifically comprises the steps of:

[0149] generating navigation instructions according to the location information;

[0150] The navigation instruction is executed to realize processing of the target object.

[0151] In specific implementation, the technical effect that can be achieved by this embodiment is to build an automated object recognition and processing system, which first uses a large language model to perform in-depth analysis on the feature description of the object and accurately identify the target object; then, based on the recognition result, the precise position of the target object in the target image is determined, and necessary navigation instructions are provided to the robot or other automated equipment; finally, these navigation instructions are executed to complete the precise manipulation of the target object, such as grabbing, carrying or placing. The entire process realizes a fully automated process from image analysis to object manipulation, improves the efficiency and accuracy of the operation, reduces manual intervention, and is suitable for multiple fields such as automated production lines, intelligent monitoring systems, and robot vision applications.

[0152] The specific implementation process of the steps in this embodiment is roughly as follows:

[0153] 1. Build a large language model:

[0154] Model selection: Choose a suitable pre-trained language model, such as BERT, GPT, etc., which can understand and process natural language text;

[0155] Model fine-tuning: For a specific object recognition task, use relevant data to fine-tune the model to improve its accuracy on the current task;

[0156] 2. Feature description input and recognition result generation:

[0157] Model reasoning: The feature description of each object is input into the large language model, which generates the recognition result of the target object by analyzing the text;

[0158] 3. Object position calculation:

[0159] Result analysis: Extract the recognition result of the target object from the output of the large language model, and calculate the location information of the target object in the target image in combination with the metadata of the target image;

[0160] Coordinate determination: Determine the bounding box coordinates of the target object or the key point position of the target object to provide the necessary spatial information for navigation instructions;

[0161] 4. Navigation instruction generation:

[0162] Path planning: Plan the path from the current position to the target object based on the target object’s location information;

[0163] Instruction encoding: Encode the planning results into navigation instructions, which can be moving direction, distance or specific action commands;

[0164] 5. Execute actions to process the target object:

[0165] Action decision: Based on the navigation instructions and the recognition results of the target object, the specific action to be performed is determined, such as grasping, moving or placing;

[0166] Action execution: robots or other automated devices perform these actions according to navigation instructions and handle target objects;

[0167] 6. Feedback and optimization:

[0168] Execution monitoring: monitor the process of action execution to ensure that the action is executed correctly according to the navigation instructions;

[0169] Performance feedback: Based on the feedback from the execution results, the relevant models and navigation algorithms are optimized to improve the accuracy and efficiency of the object recognition system;

[0170] Through the above process, this embodiment can build an automated process from object recognition to object processing, thereby achieving efficient and accurate processing of target objects.

[0171] It can be seen from the above method embodiments that the object recognition method provided by the present invention includes: determining the target object to be recognized, and using an image acquisition device to collect an image containing the target object in real time; preprocessing the collected image to obtain a target image; extracting all objects in the target image using an image segmentation algorithm; using a large visual model to perform feature recognition on each of the objects, and converting the recognized features into feature descriptions; based on the feature descriptions of each of the objects, generating the recognition result of the target object through a large language model. In this way, the method of the present invention can solve the problem that the current object recognition method in the prior art has deficiencies in accuracy, robustness, and scalability.

[0172] It can be understood that the object recognition method provided by the embodiment of the present invention can be applied to object recognition scenarios related to the medical and health field. Specifically, the object recognition function can be integrated into the corresponding AI intelligent accompanying robot of the medical institution. In the field of medical accompanying, the application scenarios of AI intelligent accompanying robots are wide, including but not limited to ward rounds, patient monitoring, drug distribution, assisted rehabilitation training, etc.; through the object recognition method provided by the embodiment of the present invention, these AI intelligent accompanying robots can identify and process specific items in the medical environment, such as drugs, medical equipment, etc. through precise object recognition capabilities.

[0173] It can be understood that the object recognition method provided by the embodiment of the present invention not only improves the accuracy and robustness of object recognition, but also achieves seamless integration with the AI ​​intelligent accompanying robot, enabling it to perform complex tasks in the medical environment, such as drug delivery, patient monitoring and assisted rehabilitation training, etc.; the application of the method of the present invention will greatly improve the quality and efficiency of medical accompanying, while reducing the workload of medical staff, enabling them to devote more time and energy to medical activities that require more professional skills; at the same time, for patients, accurate object recognition capabilities mean more personalized and timely nursing services, which improves patient satisfaction and rehabilitation effects.

[0174] In short, the implementation of the object recognition method provided by the embodiment of the present invention has important practical significance for promoting the intelligent and automated development of the medical and health field. It can not only improve the level of medical services, but also promote the innovation and application of related technologies, bringing far-reaching economic and social benefits to society.

[0175] It should be understood that, although the present application provides method operation steps as described in the embodiments or flowcharts, more or less operation steps may be included based on conventional or non-creative labor, and these operation steps are not necessarily performed in sequence according to the embodiment or flowchart. The order of steps listed in the embodiment or flowchart is only one way of executing the order of many steps, and does not represent the only execution order. It should be noted that there is not necessarily a certain order between the above steps. A person of ordinary skill in the art can understand from the description of the embodiment of the present invention that in different embodiments, the above steps may have different execution orders, that is, they may be executed in parallel, or they may be executed in exchange, etc. Moreover, at least a part of the steps in the embodiment or flowchart may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but may be executed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but may be executed in turn, alternately or synchronously with other steps or at least a part of the sub-steps or stages of other steps.

[0176] Based on the above method embodiment, please refer to Figure 3 Another embodiment of the present invention further provides an object recognition device, wherein the device comprises:

[0177] The acquisition module 11 is used to determine the target object to be identified and to acquire an image containing the target object in real time using an image acquisition device;

[0178] A preprocessing module 12 is used to preprocess the collected image to obtain a target image;

[0179] An extraction module 13, used to extract all objects in the target image using an image segmentation algorithm;

[0180] A conversion module 14, for performing feature recognition on each of the objects using a large visual model, and converting the recognized features into feature descriptions;

[0181] The result generating module 15 is used to generate the recognition result of the target object through a large language model based on the feature description of each object.

[0182] Furthermore, in one embodiment, in the object recognition device, the acquisition module 11 is specifically used for:

[0183] Determine the target object to be identified according to the task objective;

[0184] Determine the image acquisition device based on the task environment;

[0185] The image acquisition device is used to acquire images containing the target object in real time.

[0186] Furthermore, in one embodiment, in the object recognition device, the preprocessing module 12 is specifically used for:

[0187] After performing image denoising, image enhancement and image conversion on the collected image, a preliminary image is obtained;

[0188] The preliminary image is subjected to quality inspection, and when the inspection result meets the preset image quality requirement, the target image is obtained.

[0189] Furthermore, in one embodiment, in the object recognition device, the extraction module 13 is specifically used for:

[0190] Determining an image segmentation algorithm according to image content features of the target image;

[0191] The image segmentation algorithm is used to extract all objects in the target image, and a corresponding unique identifier is assigned to each of the extracted objects.

[0192] Furthermore, in one embodiment, in the object recognition device, the conversion module 14 is specifically configured to:

[0193] Pre-build visual models;

[0194] Performing feature recognition on each of the objects through the visual large model, and converting the recognized features of each of the objects into feature text;

[0195] The feature text is sorted to generate a feature description.

[0196] Furthermore, in one embodiment, in the object recognition device, the result generation module 15 is specifically used for:

[0197] Pre-build large language models;

[0198] Inputting the feature description of each of the objects into the large language model to generate a recognition result of the target object;

[0199] Based on the recognition result, position information of the target object in the target image is determined.

[0200] Furthermore, in the object recognition device, the result generation module 15 is further specifically used for:

[0201] generating navigation instructions according to the location information;

[0202] The navigation instruction is executed to realize processing of the target object.

[0203] It should be noted that in the embodiment of the device of the present invention, the information interaction, execution process and other contents between the above-mentioned modules are based on the same concept as the embodiment of the method of the present invention. Their specific functions and technical effects can be found in the aforementioned method embodiment part and will not be repeated here.

[0204] Based on the above method embodiment, another embodiment of the present invention further provides a computer device, which may be a server, and its internal structure diagram may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, the functions or steps of the object recognition method server side in any of the above method embodiments are implemented.

[0205] Based on the above method embodiment, another embodiment of the present invention further provides a computer device, which may be a client, and its internal structure diagram may be as follows: Figure 5As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, the functions or steps of the client side of the object recognition method in any of the above method embodiments are implemented.

[0206] Those skilled in the art will understand that Figure 4 and Figure 5 The structural schematic diagram shown in the figure is only a schematic diagram of a partial structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied. The specific computer device may include more components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0207] The processor may be a CPU, or other general-purpose processors, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor, or the processor may be any conventional processor, etc.

[0208] Among them, the memory includes a readable storage medium, an internal memory, etc., wherein the internal memory can be the memory of a computer device, and the internal memory provides an environment for the operation of an operating system and computer-readable instructions in the readable storage medium. The readable storage medium can be a hard disk of a computer device, and in other embodiments, it can also be an external storage device of a computer device, for example, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on a computer device. Further, the memory can also include both an internal storage unit of a computer device and an external storage device. The memory is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as the program code of a computer program, etc. The memory can also be used to temporarily store data that has been output or is to be output.

[0209] Based on the above method embodiments, another embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the object recognition method in any of the above method embodiments is implemented. The computer-readable storage medium may be non-volatile or volatile.

[0210] It should be noted that the above-mentioned functions or steps that can be implemented by the computer-readable storage medium or computer device, and the technical effects brought about by the functions / steps, can be found in the relevant descriptions in the aforementioned method embodiments. To avoid repetition, they will not be described one by one here.

[0211] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM). The disclosed memory components or memories of the operating environments described herein are intended to comprise one or more of these and / or any other suitable types of memory.

[0212] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, in the embodiment of the device of the present invention, only the division of the above-mentioned functional units and modules is used as an example. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention. The specific working process of the units and modules in the above-mentioned device can refer to the corresponding process in the above-mentioned method embodiment, which will not be repeated here. If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium.

[0213] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0214] In the embodiments provided by the present invention, it should be understood that the disclosed devices / computer equipment and methods can be implemented in other ways. For example, the device / computer equipment embodiments described above are only schematic, for example, the division of modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0215] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0216] It should be noted that if software tools or components other than those of the Company appear in the embodiments of the present application, they are only used for illustration and do not represent actual use. The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the above embodiments, a person of ordinary skill in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features thereof can be replaced by equivalents; and these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A method for object recognition, characterized in that: include: Determine a target object to be identified, and use an image acquisition device to acquire an image containing the target object in real time; Preprocessing the collected image to obtain a target image; Using an image segmentation algorithm to extract all objects in the target image; Using the visual macromodel to perform feature recognition on each of the objects, and converting the recognized features into feature descriptions; Based on the feature description of each of the objects, a recognition result of the target object is generated through a large language model.

2. The object recognition method according to claim 1, characterized in that: The step of determining a target object to be identified and using an image acquisition device to acquire an image containing the target object in real time includes: Determine the target object to be identified according to the task objective; Determine the image acquisition device based on the task environment; The image acquisition device is used to acquire images containing the target object in real time.

3. The object recognition method according to claim 1, characterized in that: The preprocessing of the acquired image to obtain a target image includes: After performing image denoising, image enhancement and image conversion on the collected image, a preliminary image is obtained; The preliminary image is subjected to quality inspection, and when the inspection result meets the preset image quality requirement, the target image is obtained.

4. The object recognition method according to claim 1, characterized in that: The extracting all objects in the target image by using an image segmentation algorithm comprises: Determining an image segmentation algorithm according to image content features of the target image; The image segmentation algorithm is used to extract all objects in the target image, and a corresponding unique identifier is assigned to each of the extracted objects.

5. The object recognition method according to claim 1, characterized in that: The method of using the visual macro model to identify features of each object and converting the identified features into feature descriptions includes: Pre-build visual models; Performing feature recognition on each of the objects through the visual large model, and converting the recognized features of each of the objects into feature text; The feature text is sorted to generate a feature description.

6. The object recognition method according to any one of claims 1 to 5, characterized in that: The generating the recognition result of the target object by using a large language model based on the feature description of each object includes: Pre-build large language models; Inputting the feature description of each of the objects into the large language model to generate a recognition result of the target object; Based on the recognition result, position information of the target object in the target image is determined.

7. The object recognition method according to claim 6, characterized in that: After determining the position information of the target object in the target image based on the recognition result, the method further includes: generating navigation instructions according to the location information; The navigation instruction is executed to realize processing of the target object.

8. An object recognition device, characterized in that: include: An acquisition module, used to determine a target object to be identified, and to acquire an image containing the target object in real time using an image acquisition device; A preprocessing module, used for preprocessing the collected image to obtain a target image; An extraction module, used for extracting all objects in the target image by using an image segmentation algorithm; A conversion module, used for performing feature recognition on each of the objects using a large visual model, and converting the recognized features into feature descriptions; A result generation module is used to generate a recognition result of the target object through a large language model based on the feature description of each object.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the object recognition method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the object recognition method according to any one of claims 1 to 7 is implemented.

Images