Parasite detection system and method based on computer assistance

Abstract

The invention discloses a parasite detection system and method based on computer assistance, and relates to the field of parasite detection, and the parasite detection system comprises a control module which is used for being responsible for the coordination and management of the whole system, and enabling functional modules to work according to a preset editing sequence and logic; the detection and identification module is used for carrying out association matching on the features extracted by the factor extraction module and the features stored in the database module to obtain parasite data which can be directly identified in the current sample image; the abnormity supplementing module is used for identifying parasite data parameters of the suspected parasite and filling fuzzy missing data; according to the method, accurate matching and recognition can be carried out according to existing parasite features, unclear or incomplete parasite features can be processed through fuzzy analysis and data prediction, effective recognition can still be carried out even under the condition that the sample quality is poor, fuzzy and missing data can be actively supplemented, and researchers are helped to perfect samples.

Description

Technical Field

[0001] The present invention relates to the technical field of parasite detection, and particularly to a computer-aided parasite detection system and method. Background Art

[0002] Diseases caused by parasites pose a serious threat to human health. Especially in some developing countries, parasite infections may lead to anemia, malnutrition, and other health problems. Therefore, timely and accurate detection and diagnosis become particularly important. With the rapid development of technologies such as computer vision, image processing, and machine learning, computer-aided detection systems have become more intelligent and efficient. The progress of these technologies makes it feasible to automatically analyze and identify parasites and enables the processing of a large amount of image data. In many hospitals and laboratories, the number of professionals is limited. Traditional parasite detection methods usually require a high level of professional skills. Computer-aided systems can improve the detection efficiency and relieve the work pressure of professionals, especially in resource-scarce areas. Moreover, with the development of medical imaging technology, the parasite image database is constantly enriched. Coupled with the popularization of the Internet and cloud computing platforms, the storage, management, and sharing of data have become more convenient, which provides rich training and verification data for computer-aided detection systems. However, in traditional detection methods, due to the wide variety of parasite species and possible morphological variations, traditional detection methods may have difficulty coping with this complexity, have a weak ability to process blurred or missing images, and are prone to missing important features. Blurred or incomplete parasite samples may be difficult to correctly identify and classify. When dealing with new samples, historical detection data is often not fully utilized, and there is a lack of flexibility and adaptability in different samples or different environments. Summary of the Invention

[0003] (1) Technical problems to be solved: Aiming at the above-mentioned disadvantages of the prior art, the present invention provides a computer-aided parasite detection system and method, which can effectively solve the problems of the prior art.

[0004] (2) Technical solutions: To achieve the above object, the present invention is realized through the following technical solutions. The present invention discloses a computer-aided parasite detection system, including: A control module, which is responsible for the coordination and management of the entire system, enabling the functional modules to work according to the pre-edited order and logic. A database module, which is used to store and classify the characteristic parameters of several parasite images and historical detection data, including the characteristics of identified parasite samples and blurred and incomplete samples. An image capture module, which is used to acquire sample images, perform preprocessing, and provide real-time previews. A factor extraction module, which is used to analyze the captured sample images and extract several characteristic parameters of parasites; A detection and recognition module, which is used to associate and match the features extracted by the factor extraction module with the features stored in the database module to obtain the directly recognizable parasite data in the current sample image; An anomaly supplementation module, which is used to identify the parasite data parameters of suspected parasites and fill in the fuzzy and missing data; A summary module, which is used to integrate the parasite parameters after prediction and filling and the parasite parameters initially identified by the detection and recognition module to generate the final detection result for output.

[0005] Furthermore, sub-modules are deployed under the anomaly supplementation module. The sub-modules include: a fuzzy determination module, a model construction module, and a prediction and filling module. The fuzzy determination module and the model construction module are connected interactively through a wireless network, and the model construction module and the prediction and filling module are connected interactively through a wireless network, where: The fuzzy determination module is used to define the fuzzy or incomplete criteria, determine the fuzzy or incomplete parts through comparative analysis with clear images, analyze whether there are fuzzy or incomplete parasite features in the sample image, and mark the fuzzy or incomplete areas as the parts to be filled; The model construction module is used to train the recognition model based on the features of historical fuzzy and incomplete parasite parameters and the features of real parasite parameters; The prediction and filling module is used to extract the trained recognition model in the model construction module, input the fuzzy and incomplete parasite parameters, predict and output the parameters of the fuzzy and incomplete parasite parts, and complete the filling of the simulated incomplete parasite parameters based on the output data.

[0006] Furthermore, the fuzzy determination module is connected interactively through a wireless network with an active supplementation module. The active supplementation module is used to customize the docking with the fuzzy determination module, actively edit the fuzzy or incomplete parasite features, and edit the fuzzy or incomplete criteria.

[0007] Furthermore, the prediction and filling module is connected interactively through a wireless network to an associated traceability module. The associated traceability module is used to correspondingly retrieve the sample image data based on the prediction data output by the prediction and filling module, and provide the area to be filled of the sample image data to the prediction and filling module.

[0008] Furthermore, the summarization module is connected to a feedback verification module through wireless network interaction. The feedback verification module is connected to the database module through wireless network interaction. The feedback verification module is used to monitor and record the effect of each prediction filling, classify and store the summarized data in the database module for use as reference parameters, receive the true verification results, and correct the data according to the verification results, and update the parameters in the abnormal supplement module through user feedback.

[0009] Furthermore, the calculation formula for the prediction filling module to predict and output the fuzzy and incomplete parasite part parameters is: ; In the formula, represents the missing parameters predicted by the model based on the input data, represents the weighting factor of the prediction result, which is used to balance the reliability and fuzziness of the output, represents the number of features selected for data characteristics, represents the feature associated with the th weight coefficient, reflecting the importance of each feature in the prediction, represents the fuzzy parameter , historical data and context information as the input th feature extraction function, and outputs the feature value associated with the position of the missing parameter, represents the fuzzy and incomplete parasite parameter vector, which contains some known features and some unknown features, represents the known parasite features in the historical data, which are used for reference and training the model, represents the context information, such as more context parameters such as the type, age, and environmental factors of the parasite.

[0010] Furthermore, the calculation formula of the is: ; In the formula, represents bias term of, indicating the base value of this feature, represents the total number of fuzzy parameters of the current feature extraction function, represents the th feature extraction function associated with the th input parameter weight coefficient, represents the components of the fuzzy and incomplete parameters, extracted from , represents the weight associated with the input parameter , indicating the importance of this feature in predicting the missing parameter.

[0011] Further, the control module is interactively connected to the database module and the image capture module through a wireless network. The image capture module is interactively connected to the factor extraction module through a wireless network. The detection and recognition module is interactively connected to the factor extraction module and the anomaly supplement module through a wireless network. The anomaly supplement module is interactively connected to the summary module through a wireless network.

[0012] A computer-aided parasite detection method includes the following steps: Step 1: Obtain image data of a sample to be detected through an image capture device, and perform preprocessing to obtain a sample image. Step 2: Process the obtained sample image to extract the shape, color, and texture features of the unique parasites therein. Step 3: Compare the extracted parasite features with the parasite features stored in the database, and identify the types of parasites existing in the sample image through a matching algorithm. Step 4: Analyze the possibly blurred or incomplete parasite features in the sample image, and record the blurring or incompleteness parameters. Step 5: Compare the verified blurred and incomplete parasite features with the real parasite features in the historical data, and train a recognition model through the historical data. Step 6: For the detected blurred and incomplete parasite features, use the trained recognition model to speculate and fill in the missing or unclear parameters to generate a complete description of the parasite features. Step 7: Integrate the filled parasite parameters with the initially identified parasite features to form a comprehensive parasite detection result. Step 8: Classify the integrated detection results, and conduct real verification in a timely manner. Update and correct the relevant parameters in the database through the verification results.

[0013] Further, the recognition model in Step 6 is constructed through a deep learning algorithm, accepts the input blurred parameters and known parameters, uses the weights obtained from the model output and the feature extraction function, generates a filling output through forward propagation, and combines the calculated filling parameters with the known parameters to generate a complete parameter vector.

[0014] (III) Beneficial effects: Adopting the technical solution provided by the present invention, compared with the known prior art, the following beneficial effects are achieved. 1. Through automated image capture and feature extraction, the system can quickly process a large number of sample images, significantly improving the detection efficiency, reducing the time and effort of manual operations. Using the database and model training, the system can perform precise matching and identification based on existing parasite characteristics, reducing errors caused by human judgment. It can efficiently store, classify, and manage historical detection data for subsequent retrospective analysis and optimization, while supporting real-time monitoring and recording of detection experiments.

[0015] 2. Through fuzzy analysis and data prediction, the system can process unclear or incomplete parasite characteristics, enabling effective identification even when the sample quality is poor. It actively supplements fuzzy and missing data to help researchers improve the samples. The feedback verification mechanism allows it to continuously update and improve the database. Over time and with data accumulation, the detection accuracy and adaptability of the system can be enhanced.

[0016] 3. By efficiently storing, classifying, and managing historical detection data, it facilitates subsequent retrospective analysis and optimization, while supporting real-time monitoring and recording of detection experiments, making the management, maintenance, and update of data during the parasite detection process more efficient, providing a good foundation for subsequent research and analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0018] Figure 1 It is a schematic framework diagram of a computer-aided parasite detection system in the present invention; Figure 2 It is a schematic flowchart of a computer-aided parasite detection method in the present invention.

[0019] The reference numerals in the figure respectively represent: 1, control module; 2, database module; 3, image capture module; 4, factor extraction module; 5, detection and identification module; 6, anomaly supplementation module; 61, fuzzy determination module; 62, model construction module; 63, prediction filling module; 7, summary module; 8, feedback verification module; 9, active supplementation module; 10, associated traceability module. DETAILED DESCRIPTION OF THE INVENTION

[0020] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0021] The following further describes the present invention with reference to embodiments. Embodiment

[0022] The computer-aided parasite detection system of this embodiment, as Figure 1 shown, includes: A control module 1, which is responsible for the coordination and management of the entire system, enabling the functional modules to work in accordance with the pre-edited order and logic; ensuring that the entire system operates according to the established process; A database module 2, which is used to store and classify the characteristic parameters of several parasite images and historical detection data, including the characteristics of identified parasite samples and blurred and incomplete samples; by obtaining the information of clear samples and blurred samples, it is convenient for subsequent comparison and matching; An image capture module 3, which is used to acquire sample images, perform preprocessing, and provide real-time previews; A factor extraction module 4, which is used to analyze the captured sample images and extract several characteristic parameters of parasites; A detection and recognition module 5, which is used to associate and match the characteristics extracted by the factor extraction module 4 with the characteristics stored in the database module 2 to obtain the directly recognizable parasite data in the current sample image; An anomaly supplement module 6, which is used to identify the parasite data parameters of suspected parasites and fill in the blurred and missing data; ensure that each characteristic in the sample is deeply analyzed and completed; There are sub-modules deployed under the anomaly supplement module 6. The sub-modules include: a blur determination module 61, a model construction module 62, and a prediction and filling module 63. The blur determination module 61 and the model construction module 62 are connected through wireless network interaction, and the model construction module 62 and the prediction and filling module 63 are connected through wireless network interaction, where: A blur determination module 61, which is used to define the blur or incompleteness criteria, determine the blurred or incomplete parts through comparative analysis with clear images, analyze whether there are blurred or incomplete parasite characteristics in the sample image, and mark the blurred or incomplete areas as parts that need to be filled; The fuzzy determination module 61 is connected to the active supplement module 9 through wireless network interaction. The active supplement module 9 is used to customize the connection to the fuzzy determination module 61, actively edit the fuzzy or incomplete parasite features, and edit the fuzzy or incompleteness criteria; The model construction module 62 is used to train the recognition model based on the features of the historical fuzzy and incomplete parasite parameters and the features of the real parasite parameters; The prediction and filling module 63 is used to extract the trained recognition model in the model construction module 62, input the fuzzy and incomplete parasite parameters, predict and output the fuzzy and incomplete parasite part parameters, and complete the filling of the simulated incomplete parasite parameters based on the output data; The prediction and filling module 63 is connected to the associated traceability module 10 through wireless network interaction. The associated traceability module 10 is used to retrieve the sample image data corresponding to the prediction data output by the prediction and filling module 63, and provide the area to be filled of the sample image data to the prediction and filling module 63; The summary module 7 is used to integrate the parasite parameters after prediction and filling and the parasite parameters initially identified by the detection and recognition module 5, and generate the final detection result for output; The summary module 7 is connected to the feedback verification module 8 through wireless network interaction. The feedback verification module 8 is connected to the database module 2 through wireless network interaction. The feedback verification module 8 is used to monitor and record the effect of each prediction and filling, classify and store the summarized data in the database module 2 as reference parameters, receive the true verification result and correct the data according to the verification result, and update the parameters in the abnormal supplement module 6 through user feedback.

[0023] As an implementation mode in this embodiment, as Figure 1 shown, the control module 1 is connected to the database module 2 and the image capture module 3 through wireless network interaction. The image capture module 3 is connected to the factor extraction module 4 through wireless network interaction. The detection and recognition module 5 is connected to the factor extraction module 4 and the abnormal supplement module 6 through wireless network interaction. The abnormal supplement module 6 is connected to the summary module 7 through wireless network interaction.

[0024] In the specific implementation of this embodiment, the control module 1 controls the global function module. The database module 2 stores the characteristic factors of several parasite images. The image capture module 3 acquires sample images. The factor extraction module 4 extracts the parasite characteristic factors in the sample images. The detection and recognition module 5 matches with the stored data in the database module to identify the parasite information in the current sample image. The fuzzy determination module 61 analyzes whether there are fuzzy and incomplete parasite parameters that cannot be recognized in the sample image. Model training is performed based on the characteristics of historical fuzzy and incomplete parasite parameters and the characteristics of verified control true parasite parameters. The model construction module 62 constructs an identification model. The prediction and filling module 63 extracts the identification model. By inputting fuzzy and incomplete parasite parameters, the missing or fuzzy parasite part parameters are predicted and output to complete the filling of the simulated incomplete parasite parameters. The summary module 7 integrates and outputs the parasite parameters after prediction and filling and the parasite parameters initially identified by the detection and recognition module. In addition, through the feedback verification module 8, the integrated data is classified and stored in the database module 2 as reference parameters to receive the true experimental verification results of the predicted and filled parasite parameters. According to the verification results, the database module 2 is modified. The active supplement module 9 actively adjusts the parameters of the fuzzy determination module 61, and the associated traceability module 10 provides the associated filling objects for the fuzzy missing data. Embodiment

[0025] On other levels, this embodiment also provides a computer-aided parasite detection method, as Figure 2 shown, including the following steps: Step 1: Obtain the image data of the sample to be detected through an image capture device, perform preprocessing, and use it as a sample image; Step 2: Process the obtained sample image to extract the unique shape, color, and texture features of the parasites therein; Step 3: Compare the extracted parasite features with the parasite features stored in the database, and identify the types of parasites existing in the sample image through a matching algorithm; Step 4: Analyze the possible fuzzy or incomplete parasite features in the sample image, and record the fuzzy or incomplete parameters; Step 5: Use the verified fuzzy and incomplete parasite features and true parasite features in the historical data for comparison, and train an identification model through the historical data; Step 6: For the detected fuzzy and incomplete parasite features, use the trained identification model to speculate and fill in the missing or unclear parameters to generate a complete parasite feature description; The recognition model is constructed through deep learning algorithms. It accepts the input fuzzy parameters and known parameters, uses the weights obtained from the model output and the feature extraction function, generates a filled output through forward propagation, and combines the calculated filled parameters with the known parameters to generate a complete parameter vector; Step 7: Integrate the filled parasite parameters with the initially identified parasite features to form a comprehensive parasite detection result; Step 8: Classify the integrated detection results, and conduct real - time verification in a timely manner. Update and correct the relevant parameters in the database based on the verification results.

[0026] Compared with the prior art, automatically obtaining sample images using an image capture device and processing them with computer algorithms significantly improves the speed and efficiency of parasite detection. Compared with traditional manual detection methods, it can obtain detection results faster. By extracting sample image features, including shape, color, and texture, etc., and combining deep learning algorithms for feature comparison and matching, it can more accurately identify parasite species and reduce the situations of misdiagnosis or missed diagnosis; By particularly focusing on fuzzy or incomplete parasite features, the trained recognition model supplements and infers missing or unclear parameters, thereby improving the adaptability and recognition rate for complex samples, which is often overlooked in traditional methods; Using historical data for model training and updating and correcting the relevant parameters in the database through actual verification results enables this detection method to have the ability of self - learning. As the detected data increases, the accuracy and generality of the model are continuously improved. By integrating the filled parasite features with the initially identified results to form a comprehensive detection conclusion, information loss during the detection process is reduced, ensuring the comprehensiveness of the results; It can classify and verify detection results in real - time, timely feedback positive or negative results, improve the timeliness of clinical decision - making, and adapt to the medical needs of rapid response. Embodiment

[0027] In this embodiment, a calculation formula for predicting and outputting the parameters of a fuzzy and incomplete parasite part is provided, specifically: ; In the formula, represents the missing parameters predicted by the model based on the input data, represents the weighting factor of the prediction result, which is used to balance the reliability and fuzziness of the output, represents the number of features selected for data features, represents the feature associated with the th weight coefficient, which reflects the importance of each feature in the prediction, represents the fuzzy parameter historical data and context information is the th feature extraction function for the input, and outputs the feature values associated with the positions of the missing parameters. represents the fuzzy incomplete parasite parameter vector, which contains some known features and some unknown features. represents the known parasite features in the historical data, which are used for reference and training the model. represents the context information, such as more context parameters like the species, age, environmental factors of the parasite, etc. The calculation formula of is: represents the bias term of represents the total number of fuzzy parameters of the current feature extraction function. represents the th feature extraction function associated with the th input parameter weight coefficient of represents the th component of the fuzzy incomplete parameter, extracted from represents the weight associated with the input parameter indicating the importance of this feature in predicting the missing parameter.

[0028] can systematically derive the feature values related to the missing and fuzzy features of the parasite from the fuzzy data, historical data and context, making the final prediction specifically fill in the specific parasite features.

[0029] In summary, the present invention uses technologies such as image processing and computer vision to extract the parasite features in the image, and performs matching and prediction through machine learning algorithms to achieve the purpose of efficient recognition. By extracting and analyzing the features of the parasite and combining historical data for model training, the system can identify multiple parasites, and predict and fill in the fuzzy and incomplete features, thereby improving the recognition accuracy, being able to effectively process the sample images containing fuzzy or incomplete parasite features, and through fuzzy analysis and model prediction, realizing the available analysis of the non-ideal images. Through the feedback verification mechanism, the present invention can continuously update and optimize the parameters in the database according to the true verification results. This self-learning ability helps to improve the performance of the system in future detections. By centrally storing the parasite feature information, the management, maintenance, and update of data during the parasite detection process become more efficient, providing a good foundation for subsequent research and analysis. Analyzing the sample images in real time, quickly feedback the detection results, and predicting and filling in the fuzzy features, making the detection results more comprehensive and accurate.

[0030] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A computer-aided parasite detection system, characterized in that: include: The control module (1) is responsible for the coordination and management of the entire system, so that the functional modules can work according to the predetermined editing sequence and logic; A database module (2) is used to store and classify characteristic parameters and historical detection data of a number of parasite images, including characteristics of identified parasite samples and fuzzy and incomplete samples; An image capture module (3) is used to acquire sample images, perform preprocessing, and provide real-time preview; A factor extraction module (4) is used to analyze the captured sample image and extract several characteristic parameters of the parasite; A detection and identification module (5) is used to associate and match the features extracted by the factor extraction module (4) with the features stored in the database module (2) to obtain parasite data that can be directly identified in the current sample image; The abnormal supplement module (6) is used to identify the parasite data parameters of suspected parasites and fill in the fuzzy missing data; The summarization module (7) is used to integrate the predicted parasite parameters after filling and the parasite parameters initially identified by the detection and identification module (5) to generate a final detection result output.

2. The computer-aided parasite detection system according to claim 1, characterized in that: The abnormal supplement module (6) is provided with submodules at the lower level, and the submodules include: a fuzzy determination module (61), a model construction module (62) and a prediction filling module (63), the fuzzy determination module (61) and the model construction module (62) are interactively connected via a wireless network, and the model construction module (62) and the prediction filling module (63) are interactively connected via a wireless network, wherein: A blur determination module (61) is used to define blur or incompleteness standards, determine blur or incomplete parts by comparing and analyzing with clear images, analyze whether there are blur or incomplete parasite features in the sample image, and mark the blur or incomplete area as a part that needs to be filled; A model building module (62) is used to train a recognition model based on the characteristics of historical fuzzy and incomplete parasite parameters and the characteristics of real parasite parameters; The prediction and filling module (63) is used to extract the trained recognition model in the model construction module (62), input the fuzzy and incomplete parasite parameters, predict and output the fuzzy and incomplete parasite part parameters, and complete the filling of the simulated incomplete parasite parameters based on the output data.

3. The computer-aided parasite detection system according to claim 2, characterized in that: The fuzzy determination module (61) is interactively connected to an active supplement module (9) via a wireless network, and the active supplement module (9) is used to customize the docking fuzzy determination module (61) and actively edit fuzzy or incomplete parasite features and fuzzy or incompleteness standards.

4. The computer-aided parasite detection system according to claim 2, characterized in that: The prediction filling module (63) is interactively connected to the association tracing module (10) via a wireless network. The association tracing module (10) is used to retrieve sample image data corresponding to the prediction data output by the prediction filling module (63), obtain the area to be filled in the sample image data, and provide it to the prediction filling module (63).

5. The computer-aided parasite detection system according to claim 1, characterized in that: The summary module (7) is interactively connected to a feedback verification module (8) via a wireless network. The feedback verification module (8) is interactively connected to a database module (2) via a wireless network. The feedback verification module (8) is used to monitor and record the effect of each prediction and filling, classify and store the summarized data in the database module (2) as a reference parameter, receive real verification results and perform data correction according to the verification results, and update the parameters in the abnormal supplement module (6) through user feedback.

6. The computer-aided parasite detection system according to claim 2, characterized in that: The calculation formula for predicting and outputting the fuzzy and incomplete parasite part parameters by the prediction and filling module (63) is: ; In the formula, represents the missing parameters predicted by the model based on the input data, Represents the weighting factor of the prediction result, The number of features representing the data feature selection, Representative features The associated The weight coefficients, Fuzzy parameter , Historical Data and contextual information For the input feature extraction functions that output feature values ​​associated with the missing parameter locations, represents the fuzzy and incomplete parasite parameter vector, represents known parasite characteristics from historical data, Represents context information.

7. The computer-aided parasite detection system according to claim 6, characterized in that: Said The calculation formula is: ; In the formula, represent The bias term, Represents the total number of fuzzy parameters of the current feature extraction function, Representatives expressed Feature extraction function The related Input parameters The weight coefficient of Represents the fuzzy incomplete parameters Quantity, Representation and input parameters The associated weight.

8. The computer-aided parasite detection system according to claim 1, characterized in that: The control module (1) is interactively connected to the database module (2) and the image capture module (3) via a wireless network; the image capture module (3) is interactively connected to the factor extraction module (4) via a wireless network; the detection and identification module (5) is interactively connected to the factor extraction module (4) and the abnormality supplement module (6) via a wireless network; and the abnormality supplement module (6) is interactively connected to the summary module (7) via a wireless network.

9. A computer-aided parasite detection method, the method being an implementation method of the computer-aided parasite detection system according to any one of claims 1 to 8, characterized in that: The following steps are involved: Step 1: Obtain image data of the sample to be detected through an image capture device and perform preprocessing to obtain a sample image; Step 2: Process the acquired sample images to extract the shape, color and texture features of the unique parasites; Step 3: Compare the extracted parasite features with the parasite features stored in the database, and identify the type of parasite present in the sample image through a matching algorithm; Step 4: Analyze the possible blurred or incomplete parasite features in the sample image and record the blurred or incomplete parameters; Step 5: Use the verified fuzzy and incomplete parasite features in the historical data to compare with the real parasite features, and train the recognition model through the historical data; Step 6: For the fuzzy and incomplete parasite features detected, use the trained recognition model to infer and fill in the missing or unclear parameters to generate a complete parasite feature description; Step 7: Integrate the filled parasite parameters with the initially identified parasite features to form a comprehensive parasite detection result; Step 8: Classify the integrated test results and conduct real verification in a timely manner, and update and correct the relevant parameters in the database based on the verification results.

10. The computer-aided parasite detection method according to claim 9, characterized in that: The recognition model in step 6 is constructed through a deep learning algorithm, accepts input fuzzy parameters and known parameters, uses the weights obtained by the model output and the feature extraction function, generates a filled output through forward propagation, and combines the calculated output filled parameters with the known parameters to generate a complete parameter vector.

Images