RU2279187C2 - Method for receiving images - Google Patents

Abstract

FIELD: methods for receiving moving and non-moving images.

SUBSTANCE: method includes mechanical shifting of CDMA matrix with periodical return to starting point, while movement of matrix is performed in order: to right, downwards, to left, and upwards (starting position), each time moving for a value of element of matrix. Prior to each movement of matrix information (image) is read from it and recorded in memorizing device, from appropriate address. Because image is recorded serially one line after another, during sending of image to output device, no additional processing of image is required.

EFFECT: decreased error caused by movements of matrices, and also prevented reading of one and the same elements of image.

4 dwg

Description

The invention relates to methods for receiving moving and still images (digital video cameras and cameras, scanners).

A known method of receiving images using a sensor (US patent 4652928, class H 04 N 3/14, 1988), which consists in the fact that mechanical vibrations are perpendicular to the linear matrix to a linear matrix of photosensitive elements. As a result of mechanical vibrations through a horizontal scan of the transducer, each matrix element sequentially generates four samples corresponding to four image elements located in the form of a matrix within the scan period corresponding to one vibration period.

The disadvantages of the analogue method are the complexity of implementation and the achievement of image clarity only by doubling the pitch of the image element.

The closest in technical essence and the achieved result to the claimed one is the method of receiving and reproducing images (RU, patent, 2109409, class N 04 N 3/14, 1998), which consists in the fact that the matrix of photosensitive elements is mechanically displaced, the image is divided into equal windows corresponding to the number of matrix elements, the image is identified in RAM memory as a block of records, the number of records is equal to the number of pixels in one window, the matrix is shifted alternately in the horizontal direction by the window size , and in the vertical - by the size of the image element.

The disadvantages of the prototype method are the complexity of implementation, the loss of image accuracy due to errors associated with the movements of the matrix, as well as excessive reading of the same image elements.

The objective of the invention is to simplify the method and reduce errors associated with movements, as well as the exclusion of reading the same image elements.

A method of receiving images is patented, including mechanical displacement of the matrix with periodic return to the starting point, with image identification in the memory of the storage device, characterized in that the matrix is performed using the complementary metal-oxide-semiconductor technology (CMOS matrix), CMOS matrix offset carried out in the sequence: right, down, left and up to the starting position, each time moving by the value of the matrix element, and before each matrix shift, an image is taken from it and recorded in apominayuschee product at

K + 2 (N-1) + ((2M-2) + D) 2L,

where N is the number of the matrix element,

M is the row number of the matrix,

L is the number of elements in the row,

K and D depend on the sequence number of the matrix move, with K = 1, D = 0 in the initial state of the matrix, K = 2, D = 0 after the first move of the matrix, K = 2, D = 1 after the second move of the matrix, K = 1, D = 1 after the third movement of the matrix and K = 1, D = 0 after the fourth movement of the matrix and its return to its original position.

Figure 1 shows the structural diagram of a device for implementing the proposed method.

Figure 2 shows the movement scheme of the matrix.

Figure 3 presents the structure of the total image obtained as a result of applying the above method.

Figure 4 shows the recording sequence of the image obtained by applying the method in the memory of the storage device.

Below is a specific example of the implementation of the method of receiving images.

A device for implementing the method consists of a matrix made using the complementary metal-oxide-semiconductor technology 1, bus 2, a horizontal displacement piezomotor (GWP) 3, a vertical displacement piezomotor (PVP) 4, an analog-to-digital converter (ADC) 5, and a microcontroller 6, a storage device (memory) 7, a PVP cascade buffer 8, a PGP cascade buffer 9. A CMOS matrix with a resolution of 800X600 pixels was used. In the CMOS matrix, photosensitive elements are located in the X and Y plane with a step of 9.9 μm. Using GWP and PGW, the matrix is set in motion, and the pattern of its movements shown in Fig. 2 is as follows: from the starting position to the right, then down, then left and back to the starting position, each time moving by the value of the matrix element (9.9 μm). Before each movement, the image is taken from the matrix and recorded in the memory so that information from the first photosensitive cell in the first row of the matrix is recorded at address 1 in the memory, information from the second cell is recorded at 1 + 2 = 3, information from N photosensitive cells of the first row matrix is written at the address 1 + 2 (N-1). After the first shift, information from the N photosensitive cells of the first row is recorded in the memory at 2 + 2 (N-1). After the second movement of the matrix, information from the first photosensitive cell of the first row of the CMOS matrix is recorded at 2 + 2 (N-1) + 2L, where L is the number of elements in one row of the matrix. After the third movement of the matrix, information from the first photosensitive cell of the first row of the CMOS matrix is recorded at the address 1 + 2 (N-1) + 2L. The generalized formula for finding the address for any photosensitive cell of the CMOS matrix in any row and for any number of movements made is as follows:

ad = K + 2 (N-1) + ((2M-2) + D) 2L,

where ad is the address of the cell,

K is a coefficient depending on the sequence number of the movement (K = 1 - in the initial state and after the third movement, K = 2 - after the first and second movement),

N is the sequence number of the matrix element in the row,

M is the row number of the matrix,

D - coefficient, also depending on the number of movements made (D = 1 - in the initial state and after the first movement, D = 2 - after the second and third movement),

L is the number of elements in the row.

For example, the address of the third element of the CMOS matrix in the fourth row, and with two movements of the matrix in the memory, will be: ad = 2 + 2 (3-1) + ((2 × 4-2) +2) 2L = 6 + 16L.

Thus, after all the movements of the matrix and its return to its original position, the structure of the total captured image will look like that shown in figure 3, the following notation is pleasant in this figure: 1 - pixels captured when the matrix is in its original state, 2 - pixels captured after the first matrix move, 3 - pixels captured after the second matrix move, 4 - pixels captured after the third matrix move. The resolution of the total image will be 3200X2400 pixels. In the memory information is recorded sequentially line by line as shown in Fig.4.

This method, for example, will allow you to implement a video camera (camera) with a high resolution and a clear image, and the CMOS matrix itself may have a low resolution, and since this is one of the most expensive and complex devices of a digital video camera (camera), the cost of the video camera itself ( cameras) with high resolution will be significantly lower.

Since the image is recorded sequentially line by line, when this image is output to the output device (TV, monitor), additional processing of this image is not required.

Claims (1)

A method of receiving images, including the mechanical displacement of the matrix with periodic return to the starting point, with the identification of the image in the memory of the storage device, characterized in that the matrix is performed using the complementary metal-oxide-semiconductor technology (CMOS matrix), the CMOS matrix is shifted in the sequence to the right, down, left, and up to the initial position, each time moving by the value of the matrix element, and before each matrix shift, an image is taken from it and recorded in a memory triad at K + 2 (N-1) + ((2M-2) + D) 2L, where N is the number of the matrix element; M is the row number of the matrix; L is the number of elements in the row; K and D depend on the sequence number of the matrix move, with K = 1, D = 0 in the initial state of the matrix, K = 2, D = 0 after the first move of the matrix, K = 2, D = 1 after the second move of the matrix, K = 1, D = 1 after the third movement of the matrix and K = 1, D = 0 after the fourth movement of the matrix and its return to its original position.

Images