CRT color space conversion model

  • Detail

CRT color space conversion model

in the color desktop publishing system, in order to reproduce the image color more truly on the CRT, it is necessary to study the color characteristics of the CRT and realize the conversion between the CRT color space and the standard color space XYZ. Generally, when realizing the conversion between the two, it is assumed that:

(1) pressure adaptive oil source is adopted. 1) the phosphor powder has constancy and the luminous characteristics of phosphor powder do not change

2) the space is uniform, and the color development characteristics of each position on the CRT are consistent

3) the three electron guns are independent of each other. However, these assumptions are often not tenable, resulting in the imprecision of the transformation model. To solve this problem, many nonlinear transformation models are proposed. This paper analyzes and evaluates this transformation model

1. Y correction method

this method is divided into two steps. Firstly, the conversion between the brightness of CRT phosphor and the voltage of electron gun is realized; Then the conversion between CRT phosphor brightness and XYZ tristimulus value is realized

1. The relationship between CRT phosphor brightness and electron gun voltage

the relationship between CRT phosphor brightness and electron gun voltage is a nonlinear relationship, which is generally assumed to be an exponential relationship. Through analysis, it is found that when the input digital driving value Dr = DG = DB = 0, the theoretically measured three stimulus value should be equal to zero, but the actually measured three stimulus value x = 0.2327, y = 0.2201, z = 0.2956 on a CRT. In order to eliminate this error, it is necessary to subtract the measured tristimulus value when Dr = DG = DB = 0 from the measured tristimulus value, and then carry out linear regression, which can improve the calculation accuracy

Another method to solve

is to look up the table, that is, to establish the corresponding relationship table between the digital driving value and brightness value of each electron gun

2. Conversion of phosphor brightness values LR, LG, LB and tristimulus value XYZ

2. Method based on mathematical fitting

1. Linear conversion model

linear model is simple, only needs to solve 12 coefficients, and the calculation speed is fast, but the conversion accuracy is not high

2. Polynomial model

this model has high conversion accuracy and is a more practical method, but the error is large for those color conversion with color gamut compression

III. look up table method

store the measured CRT RGB and CIE XYZ standard data in the form of table. If RGB data is given during use, the corresponding XYZ data can be found from the table. Since each component of R, G and B has 256 values, the size of the table after the combination of R, G and B is 256 × two hundred and fifty-six × 256, the table is very large. For this purpose, the values of each component of R, G and B can be separated by several. If one value is taken, the table will become smaller. The data that does not exist in the table can be obtained by interpolation

table lookup method is a very simple and effective method, but the establishment of tables requires a lot of data to be tested, the workload is huge, and the tables need to occupy a lot of storage space

1. Neural network method

because the difference is too far. 1. Introduction to neural network method

neural network can complete the mapping of many nonlinear functions. It can be used to map the color in one color space to another color space, such as RGB to XYZ, HVC to XYZ, etc. In the neural network model, the multi-layer error back propagation network, namely BP network, is widely used at present. It consists of input layer, hidden layer and output layer. When realizing the conversion from Cr color space to XYZ color space, the input layer is composed of three nodes, representing R, G and B; The output layer is also composed of three nodes, corresponding to x, y and Z; For the hidden layer, it is necessary to actually determine the number of layers and nodes of the hidden layer

to realize CRT color space conversion with neural network is to learn the network according to the known input and output color data, and adjust the weight W connecting each neuron through the continuous learning of sample data, so as to make the converted color data close to the real color data. For BP network, when one input mode is given, it is transferred from the input unit to the hidden layer unit, processed layer by layer by hidden layer unit, and then sent to the output layer unit. After being processed by the output layer unit, one output mode is generated. This is a process of layer by layer state update, which is called forward network. If there is an error between the output value and the expected value and does not meet the requirements, it will turn to the error backward propagation process, transmit the error value layer by layer along the connection path, and correct the connection weight of each layer

2. Transformation model

this transformation model does not adopt the network model of 3 inputs and 3 outputs. Therefore, the network training time can be greatly reduced and the transformation accuracy can be improved

1) initialize the network learning parameters

2) provide standard color test data and train the network until the color difference is met? Br 3) forward propagation process: for a given color input data, calculate the color data of the network and compare it with the actual color data; If there is error, 20. Host weight: 1250kg; Execute 4, otherwise return 2

4) error backward propagation process

using neural network model, the color conversion accuracy is high. However, it involves many problems, such as the determination of the number of hidden layers and nodes, the setting of initial weight, the selection of learning factors and so on. The determination of these parameters needs to be carried out through experiments

methods and skills of filling ink for integrated inkjet head

inkjet printer has been loved by the majority of family users with its low price and close to the quality of laser printer, but it is troubled by the high cost of consumables at the same time. At present, manufacturers integrate the print head and ink of inkjet printer. After the ink is used up, they can only replace the print head together. It costs nearly 20 yuan, but it can only print four or five hundred pages, which is a great burden for ordinary users. In fact, after the ink is used up, the service life of the print head is not expired at all. It can be refilled with ink (ink for inkjet printer) and continue to use. Moreover, the printing quality will not change. Refilling ink is only equivalent to one quarter of the cost of replacing the print head

household inkjet printer series such as canon, Epson and HP can use the filling ink produced by a third-party manufacturer. Although the manufacturer said in the instruction manual that the failure caused by the use of filling ink is not guaranteed, there is actually no risk. If an inkjet head is used normally, there will be no problem in filling ink three or four times

when the ink is used up, carefully remove the print head from the printer and be careful not to touch the ink jet nozzle of the print head. Find a small hole of about 3mm on the side of the top of the print head with a small cylindrical plastic plug inside. Drill the small plastic plug with a small tap provided at the same time of purchasing fillable ink, pull out the small plug and put it aside. Be careful not to use moral education. Plug it after filling the ink. Cut a small hole in the long mouth of the SK filling ink bottle, insert it into the small 10 of the print head, slowly inject about 20ml of ink, be careful not to overflow, plug the small plug after filling, and then install the print head on the printer again, and it can be used normally. If you accidentally overflow the ink, you can drop it with a meal sticker. Generally, there are 30, 70, 100ml and other types of ink that can be filled. If you often print documents, you might as well buy a 100 millimetre long one, which is more suitable for the whole family

another problem that needs to be paid attention to is that the ink jet head is blocked every time after the ink is newly loaded, no matter how careful, even if it does not touch the nozzle. At this time, you must not prick the nozzle with a needle, which will completely damage the print head, because the nozzle is much smaller and delicate than we thought. Because the inkjet printer is designed based on the principle that the ink is ejected from the small hole after being heated, when the nozzle is blocked, we can immerse the part of the ink nozzle into warm water for about 1mm (only a few seconds). When the ink is heated, all the blocked inkjet nozzles will be washed away. Take out the post dinner sticker, absorb the water on the ink nozzle from the side, and then reinstall the print head, and it will be in good condition


Copyright © 2011 JIN SHI