KR19980043398A - Apparatus and method for correcting screen color according to biorhythm - Google Patents

Abstract

The present invention relates to a method of calibrating a screen color according to biorhythm using neural network theory. Conventionally, the classification of a dangerous day is treated as the state of the corresponding date and the state of the preceding and following days, and the problem of ambiguity of the body, emotion, have. Accordingly, the present invention provides a microcomputer 20 for receiving a date of birth through the key input unit 10 to calculate a biorhythm and obtaining an appropriate color correction value through a neural network, Wow; A memory unit 30 for storing a calendar table necessary for biorhythm calculation, weight values and offset values, and a color temperature value for correcting a screen color; A digital / analog conversion unit 40 for receiving the digital color temperature value from the microcomputer 20 and converting the digital color temperature value to an analog color temperature value and displaying the image signal through the chroma unit 50 to the cathode ray tube 60 The characteristic of the biorhythm curve is obtained and input to the neural network to obtain an appropriate color correction value and correct the color temperature with the color correction value, thereby making it possible to perform accurate color temperature correction.

Description

Apparatus and method for correcting screen color according to biorhythm

The present invention relates to a screen color correcting method for changing a color of a screen according to a biorhythm by receiving a date of birth of a viewer, and more particularly, to a screen color correcting method for applying a biorhythm curve to a neural network, And methods.

1, the circuit configuration of a screen color correction device according to the conventional biorhythm includes a key input unit 1 for a viewer to input a key; A microcomputer 2 that receives data input from the key input unit 1 and controls the screen color correction; A memory unit 3 for storing a reference color temperature value; A green digital / analog converter 4 for setting a color temperature value of green by the control of the microcomputer 2; A digital driving / digital converting unit 5 for driving a blue color to adjust a color temperature value of blue by the control of the microcomputer 2; And a chroma unit 6 configured to process the color temperature values set by the green and blue drive digital / analog conversion units 4 and 5 according to the control of the microcomputer 2 and display them on the cathode ray tube 7 do.

The conventional technique configured as described above will be described as follows.

When the viewer inputs his or her date of birth through the key input unit 1, the microcomputer 2 calculates the biorhythm.

The microcomputer 2 outputs 1 bit to the green / blue drive digital / analog conversion units 4 and 5, respectively, so as to output 0 bits when the rhythm is low in the calculated biorhythm.

For example, when the body rhythm of a viewer is one of the dates when the date is 0, 1, 2, 11, 12, 13, 21, 22 and the date is low, bit0 = 1 is output and the intellect rhythm is 0, 1 1, 2, 13, 14, 15, 16, 25, 26, and 27, respectively, if bit 1 is 1, 2, 15, 16, 17, 18, 19, Bit2 = 1 is output.

Therefore, the bit values of the body rhythm, intellectual rhythm, and emotional rhythm are not good (1, 1, 1).

As described above, the color temperature values according to the bit values corresponding to the body rhythm, the intellectual rhythm, and the emotional rhythm are set as shown in FIG.

The microcomputer 2 calculates the biorhythm with the date of birth inputted by the viewer and calculates the reference color temperature value stored in the memory unit 3 according to the calculated biorhythm, that is, body rhythm, intellectual rhythm, And outputs the set values to the digital / analog conversion units 4 and 5 for green and blue drive, respectively.

The green driving digital / analog converting unit 4 and the blue driving digital / analog converting unit 5 convert the digital values inputted from the microcomputer 2 into analog signals and output them to the chroma unit 6 .

The chroma unit 6 converts the analog color signals received from the green driving digital / analog converting unit 4 and the blue driving digital / analog converting unit 5 into a cathode ray tube 7).

Finally, when calculating the biorhythm of the viewer and correcting the screen color according to each calculated body rhythm, emotional rhythm, and intellectual rhythm, if the rhythm curve changes from a danger zone to a stable zone, the temperature is gradually lowered from a high color temperature to a low temperature In the stable period, when the danger zone is reached, the color temperature gradually increases to stabilize the state of mind.

However, the conventional technique as described above has a problem that the classification of a dangerous day is treated as a state of the date and a state of several days before and after, and the weights of the body, emotion, and intelligence are ambiguous.

SUMMARY OF THE INVENTION Accordingly, the present invention has been made to solve the above-mentioned problems occurring in the prior art, and it is an object of the present invention to provide an image processing apparatus and method, And a color correction apparatus and method.

BRIEF DESCRIPTION OF DRAWINGS FIG. 1 is a circuit diagram of a screen color correcting device according to a conventional biorhythm;

FIG. 2 is a diagram showing conditions for determining a color temperature according to a biorhythm in FIG. 1,

FIG. 3 is a configuration diagram of a screen color correcting device according to the biorhythm of the present invention,

Figs. 4A to 4C are graphs of biorhythm to be used for neural network learning,

4A is a graph showing an oyster rhythm, FIG. 4B is a graph showing an emotional rhythm,

Figure 4c is a graph showing body rhythm.

FIG. 5 is a flowchart illustrating a method of correcting a screen color according to the biorhythm of the present invention.

Fig. 6 is an internal configuration diagram of the memory unit in Fig. 3; Fig.

Fig. 7 is a biorhythm curve showing the date of birth of the viewer in Fig. 3; Fig.

8 is a structural view of a neural network.

9 is an input data set of each of the intelligence, emotion, and body curve input into the neural network of Fig.

10 is a graph of the input color curves input to the neural network of FIG. 8 and corresponding output color temperature graphs.

11 is a flow chart showing a training process of a neural network;

12 is a chart showing color temperature values according to neural network output;

Description of the Related Art [0002]

10: Key input unit 20: Microcomputer

30: memory unit 40: digital / analog conversion unit

50: chroma part 60: cathode ray tube

According to another aspect of the present invention, there is provided a method of correcting color of a screen according to the present invention, comprising: a first step of calculating a biorhythm by receiving a date of birth; A second step of generating neural network input data using the biorhythm calculated in the first step; A third step of providing the neural network input data generated in the second step to a neural network for training; A fourth step of determining a color temperature through a training process in the third step; And a fifth step of weighting the color temperature determined in the fourth step to determine a final color temperature to correct the color of the screen.

As shown in FIG. 3, the configuration of a screen color correction device according to the present invention includes a key input unit 10 for inputting a date of birth of a viewer, A microcomputer 20 for calculating a biorhythm by receiving the date of birth through the key input unit 10 and obtaining an appropriate color correction value again through a neural network; A memory unit 30 for storing a calendar table necessary for biorhythm calculation, weight values and offset values, and a color temperature value for correcting a screen color; And a digital / analog converter (40) for receiving a color temperature value of the digital from the microcomputer (20) and converting the color temperature value to an analog color temperature value to display a video signal through a chroma part (50) do.

The operation and effect of the present invention will be described in detail as follows.

When the viewer inputs his or her date of birth through the key input unit 10, the microcomputer 20 calculates a biorhythm corresponding to the inputted date of birth.

As shown in FIG. 4, the biorhythm is calculated by calculating an omnipotence curve having 33 periods, an emotional curve having 28 periods, and a body curve having 23 periods, and then using the calendar information table stored in the memory 30 To determine the current position (current date) of the biorhythm curve by calculating the current date. At this time, the leap year, month, etc. are distinguished, and in case of the lunar date of birth, it is converted into lift.

As described above, a magnitude of 5 days including two days before and after the current position determined in the biorhythm is applied as an input of the neural network.

Here, the reason why the input of the neural network is used for 5 days including the second day before and after the current date is to use the characteristic of calculating the shape of the input curve in the neural network.

For example, when the viewer's birthday is September 6, 1966, the microcomputer 20 calculates a biorhythm composed of intelligence, emotion, and body curves as shown in FIG. For example, if the present date is an X-day, an aggregate of X-2 days, X-1 days, X + 1 days, and X + 2 days including two days before and after the X-

That is, for each of the intelligence, emotion, and body, there are five sizes listed as {Amp (X-2), Amp (X-1), Amp (X), Amp (X + Value is input.

The intelligence, emotion, and body data output from the biorhythm are sequentially input to a neural network to obtain a color correction value through a neural network calculation process.

Here, if we look at the principle of neural networks, if input data having various forms or characteristics are repeatedly input to a neural network, the neural network outputs a single output for inputs having similar characteristics, Function.

For the neural network to distinguish between different input types, it is necessary to train the neural network itself.

In other words, iterative learning has a structure similar to that of the human brain, which can distinguish inputs according to their characteristics.

The factors that arise in this training process are weighted and offset.

Accordingly, the microcomputer 20 receives intelligence, emotion, and body input data, and inputs the weight and offset stored in the neural network parameter of the memory unit 30 to the input layer of the neural network.

The neural network is composed of an input layer for inputting data, an output layer for determining each node value for input data, and a hidden layer as an intermediate layer between the input layer and the output layer, as shown in FIG. 8, , Sensitivity, and body, respectively, and the number of nodes of the output layer is set to 20, and the number of hidden layers classified into the hidden layer is arbitrarily set to four.

And, in the neural network, the connection between the input layer and the output layer connects the neuron called the snap, and the offset is used to improve the training completion in the neural network training.

To train a neural network, various types of input data and outputs corresponding to each input are required. That is, in practice, when the input of any type comes into the test, the output from the training comes out.

The input data used in the present invention is as shown in FIG. 9, and the input data having the interval of 5 days is used and the desired output data is corresponded to each input data.

For example, if input to the neural network at the A2 input, the desired output data is

.

B2 or C2 are similar in shape but have the same output data.

When A3 is input, output data

.

By training the neural network in this way, even if there are differences in the relations of different periods, the recognition output can be made equal to the same kind of data if they are similar.

As described above, the relationship between the shape of the input curve and the corresponding output is as shown in FIG.

In the actual test, the values of the output nodes are not completely 0 or 1, and the largest value is regarded as 1 and the rest as 0.

As shown in FIG. 11, the training process of the neural network for obtaining the weights and the offsets stored in the memory unit 30 first initializes the weight wij and the offset j to an arbitrary value between 0 and 1 (S1 ).

When the above initialization process is completed, input the following equation (1) and (2).

Where Xi is the input of the input layer, Wij is the weight of the input layer i and output layer j, θj is the offset of the output layer j, and f (net j) is the output node value of the hidden layer or output layer. F (net j) is output to the hidden layer node by applying the above equation (1) between the input layer and the hidden layer.

Since the hidden layer is an input to the output layer, Equation (1) is applied to the hidden layer. At this time, the input Xi of the formula becomes f (net j) of the hidden layer node.

That is, the node of the hidden layer is an output layer for the input layer and an input layer for the output layer. Therefore, after the equation (1) is applied between the hidden layer and the output layer, an output is generated at the actual output layer node (S2).

Next, the inter-node weight Wij and the offset? J of each layer are updated (S3). This updating formula is as shown in the following equation (2).

_{W ij (t + 1) =} W ij (t) + ηδ i X i + α {W ij (t) -W ij (t-1)}

_{θ j (t + 1) =} θ j (t) + ηδ i + α {θ j (t) -θ j (t-1)} ... ... ... (2)

Here, η and δ are learning and inertia rates, which are parameters that speed up the speed of neural network training and have values ranging from 0 to 1. W (t) is the previous weight value, and W (t + 1) is the modified weight value.

After updating the weights and the offsets, the size of the error measuring the degree of training of the neural network is measured (S4).

The difference between the desired output (f (j)) and the actual output (f (net j)) is calculated and added to the error value. If this value is less than the desired error (about 0.01 information) The input data and the corresponding output data are input to the neural network and trained.

Weights and offsets are determined from the training results and used in the calculation of the neural network in the actual test.

As described above, the biorhythm according to the date of birth of the viewer is calculated in the microcomputer 20, and the color temperature values of the calculated biorhythm's intelligence, emotion, and values obtained through the neural network training process for the body are stored in the memory 30) from the color temperature table. Here, the color temperature value according to the neural network output is as shown in FIG.

In FIG. 10, since the input curve changes from the danger zone to the rising zone in FIG. 10, the color temperature is gradually lowered from the high color temperature to the low temperature. In the B zone, In this way, the color temperature is maintained at a certain level, and the color temperature is maintained at a certain level.

After determining the color temperature as above, the final color temperature is determined by multiplying the sensibility, intelligence, and body rhythm sensitive to color temperature by the appropriate influence coefficients (We, Wi, Wp) The formula to obtain is as follows.

Final color temperature = emotional color temperature × 0.5 (We) + oily color temperature × 0.3 (Wi) + body color temperature × 0.2 (Wp)

In this case, the condition of the influence coefficients (We, Wi, Wp) is set to WeWiWp.

When the final color temperature is determined through the above process, the microcomputer 20 outputs the adjustment data according to the color temperature to the digital / analog conversion unit 40.

The digital / analog conversion unit 40 converts the adjustment data according to the color temperature into an analog signal and sends the analog signal to the chroma unit 50 to adjust the R, G, and B levels to change the color temperature of the screen.

As described above, according to the present invention, the characteristics of the biorhythm curve are obtained and input to the neural network to obtain an appropriate color bubble value, and the color temperature is corrected with the color correction value, so that accurate color temperature correction can be performed.

Claims (4)

A key input unit for inputting a date of birth of a viewer; A microcomputer for receiving the date of birth through the key input unit to calculate a biorhythm and obtaining an appropriate color correction value again through a neural network; A memory unit for storing a calendar table necessary for biorhythm calculation, weight values and offset values, and a color temperature value for correcting a screen color; And a digital / analog converter for receiving the color temperature value of the digital from the microcomputer and converting the color temperature value to an analog color temperature value to display a video signal through a chroma part through a cathode ray tube. The apparatus according to claim 1, wherein the memory means comprises a calendar information table, a neural network parameter, and a color temperature table. A first step of calculating a biorhythm by receiving an input birth date; A second step of generating neural network input data using the biorhythm calculated in the first step; A third step of providing the neural network input data generated in the second step to a neural network for training; A fourth step of determining a color temperature through a training process in the third step; And a fifth step of applying a weight to the color temperature determined in the fourth step to determine a final color temperature to correct the color of the screen. [4] The method according to claim 3, wherein the final color temperature is obtained by multiplying the sensibility, the intelligibility, and the temperature of the body by their respective influence coefficients (We, Wi, Wp).