CN103745576B - The method for supervising of the display screen monitoring alarm device of child protection eyesight - Google Patents

Abstract

The present invention is a kind of display screen monitoring and alarming system of child protection eyesight, the camera comprising controller and be connected with controller, face identification unit, calculation of characteristic parameters unit, age confirmation unit, visual angle confirmation unit, time set and alarm module.Method of the present invention is mainly by judging that age, visual angle and time three conditions are monitored, and when three conditions meet, system will send warning.The invention has the beneficial effects as follows: age differentiation can be carried out to the children of viewing display screen, with strong points, do not affect the normal life of adult; Owing to achieving real-time monitoring, the impact because bad viewing custom causes children ' s vision can effectively be taken precautions against and eliminate to this system.

Description

Monitoring method of display screen monitoring alarm device for protecting children's eyesight

technical field

The invention relates to a display screen monitoring and alarming system and method for protecting children's eyesight.

Background technique

my country is one of the countries with the highest incidence of myopia in the world, and the number of myopia is the first in the world. The poor eyesight of Chinese students is very worrying. Improper sitting posture and excessive use of eyes for a long time are the main causes of myopia. Therefore, it is urgent to design a set of tools or systems to protect eyesight.

There are many kinds of eyesight protectors at home and abroad now. In China, the more commonly used one is a vision protection device installed on children's desks that can be freely retracted, and the child's sitting posture and reading and writing distance can be corrected by adjusting the height of the device. There are also some chest-mounted vision protectors, which will remind children to correct their sitting posture with a vibration alarm when their sitting posture or reading and writing distance is incorrect.

Foreign countries are also committed to the study of myopia, relatively speaking, they focus more on the causes of myopia. An American expert said that fresh vegetables and fish can protect eyesight and reduce the incidence of myopia. Through a reasonable diet and nutritional mix, the protection and prevention of eyesight can also achieve twice the result with half the effort.

The survey found that most of the devices on the market to prevent vision loss are developed for students to prevent vision loss while reading and studying. However, as electronic products become more and more common, children nowadays are quite fond of electronic products. of interest. Watching electronic products for a long time seriously threatens children's eyesight, so it is urgent to design a system for protecting children's eyesight for electronic products with display screens.

Contents of the invention

In order to solve the above technical deficiencies, the present invention provides a display screen monitoring and alarm system and method for protecting children's eyesight, which can identify and effectively monitor children watching the display screen.

The present invention is achieved through the following measures:

A display monitor and alarm system for protecting children's eyesight according to the present invention, comprising a controller and a camera connected to the controller, a face recognition unit, a characteristic parameter calculation unit, an age confirmation unit, a visual angle confirmation unit, a timing device and an alarm module , the camera is fixed on the top of the display screen.

The above-mentioned controller is connected with a display screen control switch.

The monitoring method of the display screen monitoring alarm device for protecting children's eyesight of the present invention comprises the following steps:

Step 1, collect the image in front of the display screen through the camera.

Step 2, process the image, judge whether there is a human face, if there is no human face, record the opening time t ₀ of the display screen, and judge whether it exceeds the set value T ₀ , if it exceeds T ₀ , automatically close the display screen; otherwise Return to step 1; if there is a face, go to step 3.

Step 3, extract the face feature parameters, and estimate the physiological age n of the face through analysis and processing, and judge whether the age is lower than the preset age value N, if yes, go to step 4, if not, go to step 1;

Step 4, measure the visual angle α between the eyes of the face and the display screen, and judge whether the angle α is greater than the preset angle β, if yes, proceed to step 5; if not, record the viewing time t ₁ and judge whether the time is If it exceeds the set value T ₁ , if yes, give a noise prompt, otherwise go to step 1.

Step 5, timing, if the timing time t ₂ exceeds the preset time T ₂ , send an alarm message or turn off the display; otherwise, return to step 1.

In step 3, perform wavelet transform on the extracted facial feature parameters and extract the amplitude domain map; apply the local binary mode LBP to encode the amplitude domain map, and divide it into non-overlapping rectangular blocks with the same size, calculate The histogram of each sub-block, the sequence formed by the histogram of all sub-blocks of a face image is concatenated as the texture feature vector of the face image; then the texture feature vector is reduced by principal component analysis (PCA). Dimensions, and age estimation by support vector machine SVM.

In step 2, T ₀ takes 15 minutes; in step 3, N takes 14; in step 4, β takes 45°; in step 4, T ₁ takes 5 minutes; in step 5, T ₂ takes 1 hour.

The beneficial effects of the present invention are: it is possible to judge the age of children watching the display screen, it is highly targeted, does not affect the normal life of adults, realizes real-time monitoring, effectively prevents and eliminates the impact of bad viewing habits on children's vision .

Description of drawings

Fig. 1 is the workflow block diagram of the present invention.

detailed description

Compared with the previous protective devices for protecting children's eyesight, the present invention adopts artificial intelligence processing to realize the real-time monitoring system. It is mainly aimed at electronic products with display screens, such as installing this system on TVs, notebook computers, tablet computers and smart phones, etc., which can protect the eyesight of children watching these display screens.

The system includes a controller and a camera connected to the controller, a face recognition unit, a characteristic parameter calculation unit, an age confirmation unit, a visual angle confirmation unit, a timing device and an alarm module, and the camera is fixed on the top of the display screen. The controller is connected with a display screen control switch.

The monitoring method of the display screen monitoring alarm device for protecting children's eyesight of the present invention comprises the following steps:

Step 1, collect the image located in front of the display screen through the camera;

Step 2. Process the image to determine whether there is a human face. If there is no human face, it means that the person is not in front of the display screen. Then record the opening time t ₀ of the display screen and judge whether it exceeds the set value T ₀ . If it exceeds T ₀ , Then the display screen is automatically turned off; otherwise, return to step 1; if there is a face, proceed to step 3.

Step 3, extract the face feature parameters, the feature parameter calculation module and the age confirmation module estimate the physiological age n of the face through analysis and processing, and judge whether the age is lower than the preset age value N, if yes, it means that the viewer If it is a child, proceed to step 4; if not, it means that the viewer is an adult, and return to step 1 to continue collecting images.

Step 4. Measure the visual angle α between the eyes of the face and the display screen, and judge whether the angle α is greater than the preset angle β. If yes, it means that the child is looking directly at the display screen at this time, and then proceed to step 5; At this time, the child is viewing the display screen obliquely, record the time t ₁ of viewing the display screen obliquely and judge whether the time exceeds the set value T ₁ , if so, give a noise prompt, the purpose is to let the child return to the state of viewing the display screen squarely , otherwise go to step 1 and continue to record time t ₁ .

It means that the child is viewing the display screen at an angle at this time, and proceed to step 4. If not, it means that the child is not watching the display screen, and returns to step 1 to continue collecting images;

Step 5: Time the viewing time. If the timing time t ₂ exceeds the preset time T ₂ , it means that the child has watched for too long, then send an alarm message or turn off the display screen; otherwise, return to step 1 to continue collecting images and record the viewing time t ₂ .

In step 3, the main process is to perform wavelet transform on the extracted face feature parameters and extract the amplitude domain map; apply the local binary mode LBP to encode the amplitude domain map, and divide it into non-overlapping and same size The rectangular block of each sub-block is calculated, and the sequence formed by the histogram of all sub-blocks of a face image is concatenated as the texture feature vector of the face image; then the texture feature vector is analyzed by PCA through principal component analysis. Dimensions are carried out, and the age is estimated by the support vector machine SVM.

Generally, N takes 14; in step 2, T ₀ takes 15 minutes; in step 3; in step 4, β takes 45°; in step 4, T ₁ takes 5 minutes; in step 5, Take ₁ hour for T2.

The above descriptions are only preferred embodiments of this patent. It should be pointed out that those skilled in the art can make some improvements and substitutions without departing from the technical principle of this patent, and these improvements and substitutions should also be regarded as the protection scope of this patent.

Claims (3)

1. a method for supervising for the display screen monitoring alarm device of child protection eyesight, is characterized in that, comprise the following steps:

Step 1, is positioned at the image in display screen front by camera acquisition;

Step 2, processes image, has judged whether face, if do not have face, then records display screen opening time t0, and judges whether to exceed setting value T0, if more than T0, then automatically close display screen; Otherwise return step 1; If there is face, carry out step 3;

Step 3, extracts face characteristic parameter, and estimates the physiological age n of this face by analyzing and processing, judges that whether this age is lower than default age value N, if so, then carry out step 4, if not, then carry out step 1;

Step 4, measures the eyes of this face and the visual angle α of display screen, judges whether this angle [alpha] is greater than predetermined angle β, if so, then carry out step 5; If not, then record viewing time t1 and judge whether this time exceedes setting value T1, if so, then providing noise prompting, otherwise carry out step 1;

Step 5, carries out timing, if timing time t2 exceedes preset time T 2, sends warning message or closes display screen; Then return step 1 if not.

2. the method for supervising of the display screen monitoring alarm device of child protection eyesight according to claim 1, is characterized in that: in step 3, carries out wavelet transformation to extracting face characteristic parameter and extract amplitude domain collection of illustrative plates; Application local binary patterns LBP is to amplitude domain figure spectral encoding, and be divided into not overlapping and there is the rectangular block of formed objects, calculate the histogram of each sub-block, the sequence that the histogram of all sub-blocks of a width facial image is formed is connected in series, as the texture feature vector of facial image; Then by principal component analysis (PCA) PCA, dimensionality reduction is carried out to texture feature vector, and carry out age estimation by support vector machines.

3. the method for supervising of the display screen monitoring alarm device of child protection eyesight according to claim 1, it is characterized in that: in step 2, T0 gets 15 minutes; In step 3, N gets 14 years old; In step 4, β gets 45 °; In step 4, T1 gets 5 minutes; In steps of 5, T2 gets 1 hour.