Vision detection method, vision detection system and vision detection equipment based on gesture recognition

Abstract

The invention relates to the technical field of image processing and vision detection, in particular to a vision detection method, a vision detection system and vision detection equipment based on gesture recognition, and aims to solve the problems of low efficiency and the like of preliminary vision detection judgment performed by adopting manual assistance in the prior art. The user display module is used for displaying vision sighting marks and collecting gesture pictures of the user; the gesture and sighting mark direction matching module is used for judging whether the gesture direction of the user is matched with the direction of the current actual display sighting mark or not; the control module is used for controlling conversion of levels and sizes of sighting marks, and the termination test result output module is used for terminating a test and outputting all data and a final result of current vision detection of a user. According to the invention, by designing the gesture direction and through the synergistic effect of a plurality of modules, the vision detection work is conveniently, quickly and independently completed, and the vision detection efficiency is effectively improved.

Description

Technical Field

[0001] The present application relates to the technical field of image processing and vision detection, and in particular to a vision detection method, a vision detection system and a vision detection device based on gesture recognition. Background Art

[0002] Eyes are the sense organs of vision, and their functions are vital to life and work. Many people have not kept up with the concept of preventing myopia and using their eyes properly, and as the number of myopic people increases, people are increasingly in need of quick and convenient vision testing.

[0003] The traditional vision testing method is to set up a standard logarithmic eye chart at a certain distance from the subject, and then instruct the physician to point to the visual marks on the eye chart with an indicator stick, and judge whether the subject can see the visual marks clearly by answering or making the correct direction of the corresponding visual marks. Through simple upgrading and downgrading, the subject's vision is finally determined according to the vision level of the eye chart.

[0004] In this existing testing method, the direction of the symbols on the eye chart is fixed and cannot be changed randomly. It is possible that the subject may remember them in advance or inadvertently. In addition, this testing method is inefficient and labor-intensive, and the final test results are affected by factors such as the doctor's subjective judgment, ambient lighting, and the subject's random guesses.

[0005] The existing technical document with publication number CN103598870A provides a vision detection method based on depth image gesture recognition. The method first uses a depth data acquisition device to obtain a depth image, then uses the depth image to identify the gesture area, extracts gesture features based on the identified area, and finally recognizes the gesture of the person being tested based on the extracted gesture features. There are four gesture actions: "up, down, left, and right" to match the direction of the "E" gap displayed on the display screen for matching and judgment, and obtain the final result of the vision detection.

[0006] Although the above-mentioned prior art solutions can achieve the relevant beneficial effects through the structure of the prior art, the following defects still exist: this technology imposes limitations on the gesture posture of the person being measured when the depth data acquisition device obtains the gesture picture of the person being measured (for example, the person being measured is limited to stretch his arm as far forward as possible, diagonally upward at about 30 degrees); this technology extracts gesture features through image processing to perform gesture recognition. This method is very sensitive to changes in ambient lighting, and may affect the accuracy of gesture recognition in scenes with complex, dynamic or cluttered backgrounds. In particular, for real-time gesture tracking, the captured image may lack sufficient depth information, resulting in a decrease in recognition accuracy; this technology also does not explain the logical accuracy of the vision detection process. Summary of the invention

[0007] In response to the above-mentioned existing technical problems, the present invention provides users with a vision detection method, a vision detection system and a vision detection device based on gesture recognition, so as to improve the efficiency of vision detection and reduce the waste of labor costs.

[0008] In a first aspect, the present application provides a method for vision detection based on gesture recognition, which adopts the following technical solution: A method for vision detection based on gesture recognition, the method comprising the following steps: (1) Obtaining the test subject’s vision information; (2) Based on the preset vision test interface, the vision sight mark "E" image is displayed based on the preset sight mark display size rule and random direction, and the tester's gesture screen is displayed at the same time; (3) obtaining the tester's gesture direction information, determining the tester's gesture direction based on the gesture direction information, and matching the tester's gesture direction with the direction of the visual acuity target; (4) Based on the preset vision level upgrade and downgrade rules, the size of the vision sight mark is changed and the direction is displayed randomly; (5) Based on the preset vision test rules, determine whether to continue the test or terminate the test and output the test subject's vision test result, wherein the vision test result is the previous level of the vision level displayed when the vision test is terminated.

[0009] Furthermore, obtaining the tester's vision information includes: Based on the preset user vision information acquisition interface, two options are provided to the user, namely, known vision level and unknown vision level; among which: The known vision level is: obtaining the vision level actively input by the tester; The unknown vision level is: obtaining a vision level within a random range that is consistent with daily vision testing logic.

[0010] Furthermore, the preset vision detection interface, displaying the vision sight mark "E" image based on the preset sight mark display size and random direction rules, and displaying the tester's gesture screen at the same time includes: The preset vision test interface is divided into two windows: left and right; Obtain a vision sight mark image, and display a vision sight mark "E" icon of a size and random direction that matches a certain level within a random range under the user input level or daily vision test logic in the left window; Get the tester's gesture image and display it in the right window.

[0011] Furthermore, the acquiring of the tester's gesture direction information, determining the tester's gesture direction based on the gesture direction information, and matching the tester's gesture direction with the direction of the visual acuity target includes: Acquire the gesture direction information of the tester, identify the gesture direction of the tester based on a preset gesture recognition algorithm, and display the gesture direction information; Obtain visual acuity mark display information, and match the tester's gesture pointing with the direction of the visual acuity mark based on preset gesture pointing and visual acuity mark matching rules.

[0012] Furthermore, the size change and random direction display of the vision sight mark based on the preset vision level upgrade and downgrade rule includes: Obtaining matching information between the tester's gesture direction and the direction of the visual acuity target; When the number of times the tester's hand gesture points in the same direction as the sight mark exceeds the preset upgrade threshold, the sight mark level is upgraded and the sight mark "E" image is updated based on the preset sight mark display size rule and random direction display; When the number of times the tester's hand gesture pointing is inconsistent with the direction of the vision target exceeds a preset degradation threshold, the vision target grade is lowered and the vision target "E" image is updated based on the preset vision target display size rule and random direction display.

[0013] Furthermore, the method of determining whether to continue the test or terminate the test and outputting the tester's vision test result based on a preset vision test rule, wherein the vision test result is a level before the vision level displayed when the vision test is terminated, further includes: The vision test result is output and displayed to the tester.

[0014] In a second aspect, the present application provides a vision detection system based on gesture recognition, and the technical solution adopted is as follows: the system comprises: User vision information collection module: In order to facilitate the rapid and convenient automatic vision detection, Provides a method for users to input both known vision levels and unknown vision levels; User input error reporting module: If the user input does not meet the preset rules, the user will be prompted to input incorrectly and re-enter the input; Control image and user display module: Control the display screen to divide the video window into two parts. The left side is the visual acuity sight mark display window, which is used to display the sight mark "E" image in four random directions of up, down, left, and right. The right side is the tester's gesture screen display window, which captures the user's gesture direction information for matching the gesture with the sight mark direction. Gesture and sight mark direction matching module: It uses pre-designed matching rules to determine whether the user's gesture matches the direction of the sight mark; Vision level adjustment module: Vision level adjustment is performed based on preset vision level adjustment rules and the number of correct or incorrect matches between gestures and the direction of the sight mark. Termination test result output module: Based on the preset vision test rules, terminate the vision test and process all the data of the user's vision test, and output the final test results to the tester.

[0015] In a third aspect, the present application provides a vision detection device based on gesture recognition, which adopts the following technical solution: The vision detection device includes a processor, a memory, a display screen, a camera, a keyboard, and a mouse. The memory is used to store processor executable commands, and the processor is configured to implement a vision detection method based on gesture recognition as described in any one of the first aspects.

[0016] Compared with the prior art, this application: Obtaining the user's vision information and providing the user with two options helps to improve the efficiency of vision testing. The size of the sight mark is changed and the direction is displayed randomly according to preset rules. The tester's gesture pointing is automatically recognized and matched with the direction of the sight mark. The screen is updated in real time according to the test process, which reduces the degree of manual participation in the vision testing process, realizes automated and intelligent vision testing, and outputs the final results of the vision test to the tester more quickly and conveniently.

[0017] In summary, the beneficial technical effects of this application include: (1) Improve detection accuracy: By using Media Pipe to accurately locate 21 key points of the hand, the conventional gestures used in vision detection are classified and analyzed to construct an accurate recognition algorithm of "up, down, left, and right", thereby improving the accuracy of vision detection.

[0018] (2) Realize intelligent vision testing: The person being tested can perform vision testing quickly and conveniently by themselves. For example, at home, only a computer equipped with a camera is needed to perform a quick and convenient intelligent vision test.

[0019] (3) This application is based on the traditional offline vision detection logic and optimizes it to make the entire vision detection logic process more rigorous and accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] 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 description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work in advance.

[0021] FIG1 is a method flow chart of a method for vision detection based on gesture recognition shown in an embodiment of the present application; FIG2 is an architecture diagram of a vision detection system shown in an embodiment of the present application; FIG3 is a schematic diagram of the structure of a vision detection device shown in an embodiment of the present application.

[0022] Explanation of the reference numerals: 1. Tester's vision information collection module; 2. Visual mark and tester screen display module; 3. Gesture and visual mark direction matching module; 4. Vision level improvement module; 5. Test termination and result output module. DETAILED DESCRIPTION

[0023] In order to describe the technical solutions in the embodiments of the present invention more clearly and completely, the present application is further described in detail below in conjunction with the accompanying drawings and specific embodiments. Obviously, the described embodiments are part of the embodiments of the present application, rather than 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.

[0024] The embodiment of the present application provides a method for vision detection based on gesture recognition. Figure 1 The method flow chart of a method for vision detection based on gesture recognition is described in detail, and the content includes the following steps: S1, obtaining the visual level information of the tester.

[0025] It should be noted that in order to improve the efficiency of vision testing, before starting the vision test, the tester's vision level information is first obtained, and the tester's vision level information is used to match the initial displayed sight mark level and sight mark size.

[0026] In one embodiment, the tester turns on the vision testing device and enters the tester's vision level acquisition interface. The tester selects two buttons, namely, known vision level and unknown vision level, according to the prompts. If the tester has done a vision test before and knows his or her vision level, he or she clicks the "known vision level" button and actively enters his or her vision level information. When the tester is doing a vision test for the first time or has forgotten his or her vision level, he or she clicks the "unknown vision level" button, and the system will randomly assign a vision level as the tester's vision level within a logical vision level range.

[0027] S2, enter the vision test interface, display the vision sight mark "E" image and the tester's gesture screen.

[0028] It should be noted that in S1, if the tester selects the "Known Vision Level" button, after actively entering the vision level information, click the preset "Start Test" button to enter the vision test interface; if the tester selects the "Unknown Vision Level" button, it will directly enter the vision test interface. The size of the vision sight mark "E" image is matched and displayed according to the preset rules.

[0029] In one embodiment, when the user enters the vision detection interface, the vision detection interface is divided into two display windows, the left window is a vision sight mark display screen, which is used to display the size of the vision sight mark that matches the current tester's vision level, and its direction is random, and the right window is a tester's gesture display screen, and dynamic gesture image collection begins. Further, the tester's gesture pointing image can be collected in real time and analyzed and processed.

[0030] S3, obtaining the tester's gesture pointing information, performing matching judgment with the visual mark direction based on the preset gesture pointing recognition rule, and judging whether the tester's answer is correct or not.

[0031] It should be noted that the tester's gesture pointing information is obtained based on a preset gesture pointing recognition rule, and the gesture pointing information is mainly divided into five types: "up", "down", "left", "right", and "unclear". During the vision test, if the tester is blurred or cannot see the visual mark clearly, a gesture is provided to actively tell the tester that he cannot see the currently displayed visual mark clearly. This can reduce the misjudgment caused by the tester guessing the direction of the visual mark during the vision test, and improve the accuracy of the vision test. In this application, the matching of gesture pointing and visual mark direction is a dynamic process.

[0032] In one embodiment, after the tester's gesture image is obtained, the tester's gesture direction is analyzed based on the preset gesture recognition pointing rules, the tester's gesture direction information is displayed and continuously matched with the direction of the visual mark to determine whether it is correct or not.

[0033] S4, upgrade or downgrade vision level.

[0034] It should be noted that before upgrading or downgrading the vision level based on the vision level upgrading or downgrading rules, the vision testing device will determine whether the vision at this time is the minimum vision level 4.0 and the maximum vision level 5.3. If so, there is no need to upgrade or downgrade and the results will be directly output to the tester.

[0035] The vision testing equipment sets a certain upgrade threshold to upgrade the vision level. The upgrade threshold represents the number of times the direction of the tester's hand gesture indication matches the direction of the vision target correctly, and the same applies to downgrades. Therefore, the upgrade threshold or downgrade threshold is a reasonable integer. It can be understood that when testing a certain vision level, when the number of correct matches between the tester's hand gesture pointing and the direction of the current vision target reaches the upgrade threshold, the vision level will be upgraded, and the number of correct answers to the current vision level will be increased by one. Similarly, when the number of incorrect matches between the tester's hand gesture pointing and the direction of the current vision target reaches the downgrade threshold, the vision level will be reduced, and the number of incorrect answers to the current vision level will be increased by one.

[0036] In one embodiment, the tester's gesture screen captures the tester's gesture pointing information in real time to obtain the gesture pointing of the tester's answer, and dynamically and continuously matches it with the direction of the vision target to determine whether the tester's answer is correct, that is, whether the tester can clearly see the vision target of the current vision level. When the continuous matching judgment reaches the upgrade or downgrade threshold, the vision detection device will record the number of correct or incorrect answers for the current vision level plus one, and upgrade or downgrade the vision level, and update the display interface of the vision target in S2 at the same time. The above processing can be understood as a cyclically executed, progressive vision detection algorithm.

[0037] S5, terminate the test and output the vision result to the tester.

[0038] It should be noted that the termination of vision testing depends on whether the preset termination test rules are triggered, where the vision testing termination rules include: When the vision level is 4.0, which is the lowest vision level, the tester still answers incorrectly; when the vision level is 5.3, which is the highest vision level, the tester still answers correctly. The vision test is terminated and the test results are directly output to the tester.

[0039] When a tester gives two correct answers at a certain vision level and two incorrect answers at a vision level one level higher than this vision level, the output is terminated and the vision level value of the previous level is output.

[0040] When the tester gives two wrong answers at a certain vision level during the test, and gives two correct answers at a vision level one level lower than this vision level, the output is terminated and the current vision level is output.

[0041] In one embodiment, during the vision test process, if a preset termination test rule is triggered, the test result is output to the tester and the vision test window is closed.

[0042] Based on the same technical concept, an embodiment of the present invention further provides a vision detection system, the vision detection system comprising: A test subject's vision information collection module (1) is used to obtain the test subject's vision information; An image and tester screen display module (2) is used to display the vision sight mark "E" image based on a preset sight mark display size rule and random direction, as well as a tester gesture screen display; A gesture and sight mark direction matching module (3) is used to obtain the gesture direction information of the tester, determine the gesture direction of the tester based on the gesture direction information, and match the gesture direction of the tester with the sight mark direction; A vision level upgrading module (4), used for upgrading or lowering the vision level based on preset vision level upgrading or lowering rules; The test termination result output module (5) is used to terminate the vision test based on a preset test termination rule and output the test subject's vision test result, wherein the vision test result is the previous level of the vision level when the test is terminated.

[0043] In one embodiment, the tester vision information collection module 1 further includes: The input level interface submodule is used for the tester to actively input his / her own vision level as the vision level to be tested for the initial vision test; Passively let the vision test system use a certain vision level as the vision level of the tester when first entering the vision test interface.

[0044] In one embodiment, the image and tester screen display module 2 further includes: The current vision level display submodule is used to display the vision level of the current tester being tested; An image recognition submodule, used to obtain the gesture image of the tester and identify the gesture direction of the tester based on a preset gesture recognition algorithm; The visual acuity sight mark display submodule is used to display the sight mark size that matches the current visual acuity level, and the four directions of the sight mark are randomly selected.

[0045] In one embodiment, the gesture and sight mark direction matching module 3 further includes: The gesture direction real-time display submodule is used to display the real-time gesture direction information of the current tester; The continuous matching correctness submodule is used to determine the correctness or incorrectness of the number of matches between the tester's real-time gesture pointing and the direction of the visual mark.

[0046] An embodiment of the present application discloses a vision detection device, including a memory, a processor, a camera, and a display screen. The memory stores a computer program that can be loaded by the processor and execute a vision detection method based on gesture recognition as described above.

[0047] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here.

[0048] As described above, the above embodiments are only used to introduce the technical solution of the present application in detail, but the description of the above embodiments is only used to help understand the method and core idea of ​​the present application and should not be understood as limiting the present application. Any changes or substitutions that can be easily thought of by technicians in this technical field within the technical scope disclosed in the present application should be included in the protection scope of the present application.

Claims

1. A method for vision detection based on gesture recognition, characterized in that: The method comprises the following steps: (1) Obtaining the test subject’s vision information; (2) Based on the preset vision test interface, the vision sight mark "E" image is displayed based on the preset sight mark display size rule and random direction, and the tester's gesture screen is displayed at the same time; (3) obtaining the tester's gesture direction information, determining the tester's gesture direction based on the gesture direction information, and matching the tester's gesture direction with the direction of the visual acuity target; (4) Based on the preset vision level upgrade and downgrade rules, the size of the vision sight mark is changed and the direction is displayed randomly; (5) Based on the preset vision test rules, determine whether to continue the test or terminate the test and output the test subject's vision test result, wherein the vision test result is the previous level of the vision level displayed when the vision test is terminated.

2. The method for visual acuity detection based on gesture recognition according to claim 1, characterized in that: The obtaining of the tester's vision information comprises: Based on the preset user vision information acquisition interface, two options are provided to the user, namely, known vision level and unknown vision level; among which: The known vision level is: obtaining the vision level actively input by the tester; The unknown vision level is: obtaining a vision level within a random range that is consistent with daily vision testing logic.

3. The method for visual acuity detection based on gesture recognition according to claim 1, characterized in that: The preset vision detection interface, based on which the image of the vision sight mark "E" is displayed based on the preset sight mark display size and random direction rules, and the tester's gesture screen is displayed at the same time, includes: The preset vision test interface is divided into two windows: left and right; Obtain the visual acuity sight mark image, and display the visual acuity sight mark "E" icon with a sight mark size and random direction that matches a certain level within a random range under the daily vision test logic in the left window; Get the tester's gesture image and display it in the right window.

4. The method for visual acuity detection based on gesture recognition according to claim 1, characterized in that: The step of obtaining the gesture pointing information of the tester, determining the gesture pointing of the tester based on the gesture pointing information, and matching the gesture pointing of the tester with the direction of the visual acuity target includes: Acquire the gesture direction information of the tester, identify the gesture direction of the tester based on a preset gesture recognition algorithm, and display the gesture direction information; The visual acuity mark display information is obtained, and the tester's gesture pointing is matched with the direction of the visual acuity mark based on a preset gesture pointing and visual acuity mark matching rule.

5. The method for visual acuity detection based on gesture recognition according to claim 1, characterized in that: The step of changing the size of the sight mark and displaying it in a random direction based on the preset vision level upgrade and downgrade rules includes: Obtaining the matching information between the tester's gesture direction and the visual target direction; When the number of times the tester's hand gesture points in the same direction as the sight mark exceeds the preset upgrade threshold, the sight mark level is upgraded and the sight mark "E" image is updated based on the preset sight mark display size rule and random direction display; When the number of times the tester's hand gesture pointing is inconsistent with the direction of the vision target exceeds the preset degradation threshold, the vision target level is downgraded and the vision target "E" image is updated based on the preset vision target display size rule and random direction display.

6. The method for visual acuity detection based on gesture recognition according to claim 1, characterized in that: The method of determining whether to continue the test or terminate the test based on the preset vision test rules and outputting the tester's vision test result, wherein the vision test result is the previous level of the vision level displayed when the vision test is terminated, further includes: The vision test result is output and displayed to the tester.

7. A vision detection system based on gesture recognition, characterized in that: The system comprises: A test subject's vision information collection module (1) is used to obtain the test subject's vision information; An image and user screen display module (2) is used to display the visual acuity sight mark "E" image based on a preset sight mark display size rule and random direction, as well as a tester's gesture screen display; A gesture and sight mark direction matching module (3) is used to obtain the gesture direction information of the tester, determine the gesture direction of the tester based on the gesture direction information, and match the gesture direction of the tester with the sight mark direction; A vision level upgrading module (4), used for upgrading or lowering the vision level based on preset vision level upgrading or lowering rules; The test termination result output module (5) is used to terminate the vision test based on a preset test termination rule and output the test subject's vision test result, wherein the vision test result is the previous level of the vision level when the test is terminated.

8. A vision detection device based on gesture recognition, characterized in that: The vision detection device includes a processor, a memory, a display screen, a camera, a keyboard, and a mouse. The memory is used to store processor executable commands. The processor is configured to implement a vision detection method based on gesture recognition according to any one of claims 1 to 6.

Images