Interactive vision detection system

Abstract

The interactive vision detection system comprises a terminal processor, the output end of the terminal processor is bidirectionally and electrically connected with a light source shooting system, the output end of the light source shooting system is bidirectionally and electrically connected with an image enhancement system, and the output end of the image enhancement system is bidirectionally and electrically connected with a convolutional neural network. According to the invention, the image preprocessing module preprocesses the collected original image so as to eliminate or reduce the influence of adverse factors such as noise and distortion, and then sends the processed image data to the interior of the convolutional neural network for dimension reduction processing, and the convolutional neural network fully learns the data features, so that the accuracy of dimension reduction is improved. According to the visual acuity detection system and the visual acuity detection method, the structural functionality of an existing detection system can be improved, judgment on the visual acuity detection result of the detected person through a positioning device is replaced, the phenomenon that errors occur due to positioning errors in the detection process is prevented, and the detection effect is prevented from being affected.

Description

Technical Field

[0001] The present invention relates to the technical field of vision detection systems, and particularly to an interactive vision detection system. Background Art

[0002] A vision detection system is a tool for detecting and evaluating an individual's vision condition. These systems employ advanced image processing technologies and intelligent algorithms, and can complete a comprehensive vision screening in a short time, helping individuals detect potential vision problems at an early stage. Vision detection systems usually evaluate vision through specific devices and technologies. For example, some systems use a light source system to emit light, which is focused on the fundus through a spectroscope, and then an analysis algorithm is used to evaluate various indicators of the eye. These systems can provide accurate evaluations of vision problems such as hyperopia, myopia, and astigmatism.

[0003] For example, the patent number disclosed on the Chinese Patent Network is: 201821620271.7, and the patent name is: Vision Detection System, which includes a vision chart, an indicating rod communicatively connected to the vision chart, and a server. The vision chart includes a positioning device and a circuit board. The positioning device determines the position information corresponding to the visual target on the character screen indicated by the indicating rod, and sends the position information to the circuit board. The indicating rod is provided with a confirmation button for triggering the indicating rod to send a signal to the circuit board under the action of an external force, so that when the circuit board receives the signal, it records the position information received from the positioning device at the current moment, and determines the vision detection result of the subject according to the recorded position information. The circuit board sends the vision detection result to the server. Since the position is located by the positioning device and the vision detection result is determined according to the positioning information, the probability of operator error can be reduced, and the accuracy of vision detection can be improved; at the same time, by sending the vision detection result to the server, it is beneficial for data storage and analysis.

[0004] However, the structure of the existing detection system is relatively simple. It mainly judges the vision detection result of the subject through a positioning device, and it is prone to errors due to positioning errors during the detection process, resulting in the detection effect being affected.

[0005] Therefore, it is necessary to design and transform the interactive vision detection system. Summary of the Invention

[0006] To solve the problems raised in the above background art, the purpose of the present invention is to provide an interactive vision detection system, which has the advantage of improving the detection effect, and solves the problem that the structure of the existing detection system is relatively simple, mainly judges the vision detection result of the subject through a positioning device, and is prone to errors due to positioning errors during the detection process, resulting in the detection effect being affected.

[0007] To achieve the above purpose, the present invention provides the following technical solution: An interactive vision detection system, including a terminal processor; The output end of the terminal processor is bi-directionally electrically connected to a light source shooting system. The output end of the light source shooting system is bi-directionally electrically connected to an image enhancement system. The output end of the image enhancement system is bi-directionally electrically connected to a convolutional neural network. The output end of the convolutional neural network is bi-directionally electrically connected to an OHE encoding output. The output end of the OHE encoding output is bi-directionally electrically connected to an image processing system. The output end of the image processing system is bi-directionally electrically connected to a self-noise reduction encoder. The output end of the self-noise reduction encoder is bi-directionally electrically connected to a one-dimensional convolutional neural network. The output end of the one-dimensional convolutional neural network is bi-directionally electrically connected to a comparison module.

[0008] Preferably, the output end of the comparison module is electrically connected to a feedback unit. The output end of the feedback unit is electrically connected to a dialing module. The output end of the feedback unit is electrically connected to an audible and visual alarm.

[0009] Preferably, the output end of the image processing system is bi-directionally electrically connected to a Monte Carlo tree search system. The output end of the Monte Carlo tree search system is bi-directionally electrically connected to a long short-term memory network. The output end of the long short-term memory network is bi-directionally electrically connected to a value pool processing system. The output end of the value pool processing system is bi-directionally electrically connected to the input end of the one-dimensional convolutional neural network.

[0010] Preferably, the output end of the convolutional neural network is bi-directionally electrically connected to a data caching module. The output end of the data caching module is bi-directionally electrically connected to a classification storage module. The output end of the classification storage module is electrically connected to the input end of the Monte Carlo tree search system.

[0011] Preferably, the output end of the classification storage module is bi-directionally electrically connected to a data transmission module. The output end of the data transmission module is electrically connected to a cloud server.

[0012] Preferably, the image enhancement system includes an image preprocessing module. The input end of the image preprocessing module is bi-directionally electrically connected to the input end of the light source shooting system. The output end of the image preprocessing module is bi-directionally electrically connected to a feature extraction and analysis module. The output end of the feature extraction and analysis module is bi-directionally electrically connected to an image enhancement algorithm module. The output end of the image enhancement algorithm module is bi-directionally electrically connected to a parameter optimization and selection module. The output end of the parameter optimization and selection module is bi-directionally electrically connected to an image output module. The output end of the image output module is electrically connected to the input end of the convolutional neural network.

[0013] Preferably, the input end of the terminal processor is bi-directionally electrically connected to a human-computer interaction module. The human-computer interaction module consists of a touch display screen and control buttons.

[0014] Preferably, an input end of the terminal processor is electrically connected bidirectionally to a hearing impairment assistance module, which consists of a broadcast speaker and a vibration sensor.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. The present invention preprocesses the collected original image through an image preprocessing module to eliminate or reduce the influence of adverse factors such as noise and distortion, and then sends the processed image data to the inside of the convolutional neural network for dimensionality reduction processing. The convolutional neural network fully learns the data features. On the other hand, it also performs dimensionality reduction on the feature vectors output by OHE encoding. Finally, the data is mapped to a one-dimensional convolutional neural network for display. The one-dimensional convolutional neural network image data is compared with the pre-stored user eye data without defects, so as to achieve the effect of quickly detecting the eye defects of the patient. The present invention can improve the structural functionality of the existing detection system, replace the judgment of the visual acuity detection result of the subject through a positioning device, prevent mistakes due to positioning errors during the detection process, and prevent the detection effect from being affected.

[0016] 2. By setting a feedback unit, the present invention can facilitate the user to receive the inspection result and facilitate the user to master it in a timely manner.

[0017] 3. By setting a Monte Carlo tree search system, the present invention can estimate the values of different paths by simulating multiple game or decision-making processes, and gradually establish and expand the game tree during the search process. This search method enables the Monte Carlo tree search system to find a better solution during a single traversal of the search tree, avoiding the millions of trial and errors that other algorithms may require.

[0018] 4. By setting a data cache module and a classification storage module, the present invention can store data, facilitate later search, and prevent data loss.

[0019] 5. By setting a data transmission module and a cloud server, the present invention can back up data and reduce the operating pressure on the classification storage module.

[0020] 6. By setting an image enhancement system, the present invention can preprocess the collected original image to eliminate or reduce the influence of adverse factors such as noise and distortion, extract and analyze the features of the preprocessed image, and apply corresponding image enhancement algorithms according to the extracted features and analysis results.

[0021] 7. By setting a human-computer interaction module, the present invention can facilitate the user to operate and reduce the influence caused by language barriers.

[0022] 8. By providing a broadcast speaker and a vibration sensor, the present invention facilitates the operation of hearing-impaired personnel and can further prompt operation instructions. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a schematic diagram of the system of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0024] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0025] As Figure 1 shown, an interactive vision detection system provided by the present invention includes a terminal processor; The output end of the terminal processor is bidirectionally electrically connected to a light source shooting system, the output end of the light source shooting system is bidirectionally electrically connected to an image enhancement system, the output end of the image enhancement system is bidirectionally electrically connected to a convolutional neural network, the output end of the convolutional neural network is bidirectionally electrically connected to an OHE coding output, the output end of the OHE coding output is bidirectionally electrically connected to an image processing system, the output end of the image processing system is bidirectionally electrically connected to a self-denoising encoder, the output end of the self-denoising encoder is bidirectionally electrically connected to a one-dimensional convolutional neural network, and the output end of the one-dimensional convolutional neural network is bidirectionally electrically connected to a comparison module.

[0026] Referring to Figure 1 , the output end of the comparison module is electrically connected to a feedback unit, the output end of the feedback unit is electrically connected to a dialing module, and the output end of the feedback unit is electrically connected to an audible and visual alarm.

[0027] As a technical optimization solution of the present invention, by providing a feedback unit, it is convenient for users to receive inspection results and facilitates timely mastery by users.

[0028] Referring to Figure 1 , the output end of the image processing system is bidirectionally electrically connected to a Monte Carlo tree search system, the output end of the Monte Carlo tree search system is bidirectionally electrically connected to a long short-term memory network, the output end of the long short-term memory network is bidirectionally electrically connected to a value pool processing system, and the output end of the value pool processing system is bidirectionally electrically connected to the input end of the one-dimensional convolutional neural network.

[0029] As a technical optimization solution of the present invention, by setting up a Monte Carlo tree search system, it is possible to estimate the values of different paths by simulating multiple game or decision-making processes, and gradually build and expand the game tree during the search process. This search method enables the Monte Carlo tree search system to find a better solution during a single traversal of the search tree, avoiding the millions of trials and errors that other algorithms may require.

[0030] Reference Figure 1 , the output end of the convolutional neural network is bidirectionally electrically connected to a data cache module, the output end of the data cache module is bidirectionally electrically connected to a classification storage module, and the output end of the classification storage module is bidirectionally electrically connected to the input end of the Monte Carlo tree search system.

[0031] As a technical optimization solution of the present invention, by setting up a data cache module and a classification storage module, it is possible to store data, facilitate later search, and prevent data loss.

[0032] Reference Figure 1 , the output end of the classification storage module is bidirectionally electrically connected to a data transmission module, and the output end of the data transmission module is bidirectionally electrically connected to a cloud server.

[0033] As a technical optimization solution of the present invention, by setting up a data transmission module and a cloud server, it is possible to back up data and reduce the operating pressure of the classification storage module.

[0034] Reference Figure 1 , the image enhancement system includes an image preprocessing module, the input end of the image preprocessing module is bidirectionally electrically connected to the input end of the light source shooting system, the output end of the image preprocessing module is bidirectionally electrically connected to a feature extraction and analysis module, the output end of the feature extraction and analysis module is bidirectionally electrically connected to an image enhancement algorithm module, the output end of the image enhancement algorithm module is bidirectionally electrically connected to a parameter optimization and selection module, the output end of the parameter optimization and selection module is bidirectionally electrically connected to an image output module, and the output end of the image output module is bidirectionally electrically connected to the input end of the convolutional neural network.

[0035] As a technical optimization solution of the present invention, by setting up an image enhancement system, it is possible to preprocess the collected original image to eliminate or reduce the influence of adverse factors such as noise and distortion, extract and analyze the features of the preprocessed image, and apply corresponding image enhancement algorithms according to the extracted features and analysis results.

[0036] Reference Figure 1 , the input end of the terminal processor is bidirectionally electrically connected to a human-computer interaction module, and the human-computer interaction module is composed of a touch display screen and control buttons.

[0037] As a technical optimization solution of the present invention, by setting up a human-computer interaction module, it is convenient for users to operate and reduces the impact caused by language barriers.

[0038] Reference Figure 1 , the input end of the terminal processor is bidirectionally electrically connected with a hearing impairment assistance module, and the hearing impairment assistance module is composed of a broadcast speaker and a vibration sensor.

[0039] As a technical optimization solution of the present invention, by setting up a broadcast speaker and a vibration sensor, it is convenient for hearing-impaired personnel to operate and can further prompt operation instructions.

[0040] The working principle and usage process of the present invention: When in use, the terminal processor is controlled by the human-computer interaction module to send out signals. The touch display screen and control buttons can assist language-impaired personnel in using it. At the same time, the hearing impairment assistance module can give feedback to the user to avoid the impact of hearing impairment on its use. At the same time, the terminal processor controls the light source imaging system to start up. The light source imaging system emits light, which is focused on the fundus through a beam splitter and takes pictures of the surface image of the patient's eye. The taken pictures are transmitted to the image preprocessing module. The image preprocessing module preprocesses the collected original images to eliminate or reduce the influence of adverse factors such as noise and distortion, and then sends the processed image data to the inside of the convolutional neural network for dimensionality reduction processing. The convolutional neural network fully learns the data features. On the other hand, it also reduces the dimensionality of the feature vectors output by OHE encoding. Finally, the data is mapped to a one-dimensional convolutional neural network for display. The one-dimensional convolutional neural network image data is compared with the pre-stored user eye data without defects, and the change points that appear after the comparison process are judged through the image recognition system, so as to achieve the effect of quickly detecting the eye defects of the patient.

[0041] In summary: For this interactive vision detection system, the collected original images are preprocessed by the image preprocessing module to eliminate or reduce the influence of adverse factors such as noise and distortion, and then the processed image data is sent to the inside of the convolutional neural network for dimensionality reduction processing. The convolutional neural network fully learns the data features. On the other hand, it also reduces the dimensionality of the feature vectors output by OHE encoding. Finally, the data is mapped to a one-dimensional convolutional neural network for display. The one-dimensional convolutional neural network image data is compared with the pre-stored user eye data without defects, so as to achieve the effect of quickly detecting the eye defects of the patient. The present invention can improve the structural functionality of the existing detection system, replace the judgment of the vision detection results of the examinee through the positioning device, prevent mistakes due to positioning errors during the detection process, and prevent the detection effect from being affected.

[0042] It should be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprising", "including" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or elements inherent to such process, method, article or device.

[0043] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An interactive vision detection system, including a terminal processor; Characterized in that: The output end of the terminal processor is bidirectionally electrically connected to a light source shooting system, the output end of the light source shooting system is bidirectionally electrically connected to an image enhancement system, the output end of the image enhancement system is bidirectionally electrically connected to a convolutional neural network, the output end of the convolutional neural network is bidirectionally electrically connected to an OHE encoding output, the output end of the OHE encoding output is bidirectionally electrically connected to an image processing system, the output end of the image processing system is bidirectionally electrically connected to a self-noise reduction encoder, the output end of the self-noise reduction encoder is bidirectionally electrically connected to a one-dimensional convolutional neural network, and the output end of the one-dimensional convolutional neural network is bidirectionally electrically connected to a comparison module.

2. The interactive vision detection system according to claim 1, wherein: The output end of the comparison module is electrically connected to a feedback unit, the output end of the feedback unit is electrically connected to a dialing module, and the output end of the feedback unit is electrically connected to an audible and visual alarm.

3. An interactive vision detection system according to claim 1, characterized in that: The output end of the image processing system is bidirectionally electrically connected to a Monte Carlo tree search system, the output end of the Monte Carlo tree search system is bidirectionally electrically connected to a long short-term memory network, the output end of the long short-term memory network is bidirectionally electrically connected to a value pool processing system, and the output end of the value pool processing system is bidirectionally electrically connected to the input end of the one-dimensional convolutional neural network.

4. An interactive vision detection system according to claim 3, characterized in that: The output end of the convolutional neural network is bidirectionally electrically connected to a data cache module, the output end of the data cache module is bidirectionally electrically connected to a classification storage module, and the output end of the classification storage module is bidirectionally electrically connected to the input end of the Monte Carlo tree search system.

5. An interactive vision detection system according to claim 4, characterized in that: The output end of the classification storage module is bidirectionally electrically connected to a data transmission module, and the output end of the data transmission module is bidirectionally electrically connected to a cloud server.

6. An interactive vision detection system according to claim 1, characterized in that: The image enhancement system includes an image preprocessing module, the input end of the image preprocessing module is bidirectionally electrically connected to the input end of the light source shooting system, the output end of the image preprocessing module is bidirectionally electrically connected to a feature extraction and analysis module, the output end of the feature extraction and analysis module is bidirectionally electrically connected to an image enhancement algorithm module, the output end of the image enhancement algorithm module is bidirectionally electrically connected to a parameter optimization and selection module, the output end of the parameter optimization and selection module is bidirectionally electrically connected to an image output module, and the output end of the image output module is bidirectionally electrically connected to the input end of the convolutional neural network.

7. An interactive vision detection system according to claim 1, characterized in that: The input end of the terminal processor is bidirectionally electrically connected to a human-computer interaction module, and the human-computer interaction module is composed of a touch display screen and control buttons.

8. An interactive vision detection system according to claim 1, wherein: The input end of the terminal processor is bidirectionally electrically connected to a hearing impairment assistance module, and the hearing impairment assistance module is composed of a broadcast speaker and a vibration sensor.

Images