RU2356176C2 - Method for automatic correction of optoelectronic sensor blooming and device for its realisation - Google Patents

Abstract

FIELD: physics, computer engineering.

SUBSTANCE: invention is related to computer engineering and may be used for correction of distortions in image formation by optoelectronic sensor (OES) caused by blooming - parasite flow of charges into neighboring cells of OES. Technical result is achieved by the fact that in available method for provision of television images of high precision, the following operations are introduced in chamber on regular charge-coupled device - automatic definition of image areas distorted with OES blooming, correction of images by calculation of corrected luminances of distorted areas pixels with further recording of corrected luminances of pixels of image distorted areas. Device for method realisation comprises OES, ADC, memory unit (MU) and controller with appropriate functional links. Invention makes it possible to increase accuracy of correction of distortions in any type of OES. Device makes it possible to automatically correct distortions of image on software level and may be used to increase quality of photos and video records after their recording (reception).

EFFECT: increased accuracy of automatic correction of image distortions caused by blooming.

2 cl, 3 dwg

Description

The invention relates to computer technology and can be used to correct distortions in the formation of images by an optoelectronic sensor (OED) caused by blooming - parasitic flow of charges into neighboring OED cells.

A known method for correcting distortion of the video signal of a photodetector (RF patent 2178626, publ. 2002.01.20 according to the application 2000118951/09 of 2000.07.17), which consists in correcting the video signal in accordance with distortions due to the inefficiency of charge transfer, the formation of the output of the receiving CCD matrix of the signal with the corresponding distortion coefficient, and at the output of the block for generating the correction signal of the correction signal with the same distortion coefficient, constant and uniform illumination of the control CCD matrix, similar to the receiving, source, receiving the output signal by dividing the signal taken from the lj-th element of the receiving CCD matrix by the signal taken from the lj-ro element of the control CCD matrix, where l is the number of the element in the row, j is the row number.

The disadvantage of this method is the high complexity of the device that implements this method, due to the need to use a control CCD matrix and source.

A device is known as a television camera (RF patent 2199191, publ. 2003.02.20 according to the application 2001100731/09 of 2001.01.09) containing a CCD matrix, a peak detector, an ADC, a shaper of accumulation duration, a one-shot, the first and second RS-triggers, element "OR" and inverter.

The disadvantage of this device is the low accuracy of the resulting image due to a decrease in the level of the video signal to prevent distortion caused by blooming.

Closest to the proposed method is a method for producing high definition television images in a camera on conventional CCDs (RF patent 2143789, publ. 1999.12.27 according to application 98102201/09 of 1998.01.23), which consists in the fact that the CCD video signals are shifted relative to each other along the two coordinate axes, after passing the necessary processing, they are recorded in such a way that they create an information field, and after comb filtering and full aperture correction form a high-definition television video signal.

The disadvantage of this method is the low accuracy of the correction of distortions caused by the blooming of CCDs, as well as the inability to correct distortions after receiving (recording) images.

Closest to the proposed device is a device that implements a method for producing high-definition television images in a camera on conventional CCDs (RF patent 2143789, publ. 1999.12.27 according to application 98102201/09 from 1998.01.23), containing a switch, a final processing unit, a control unit , a television camera containing an optical unit, two identical channels, each of which contains a CCD, an analog-to-digital converter (ADC), a preprocessing unit, and a memory unit.

The disadvantage of this device is the high complexity of the device, due to the need to use two CCD matrices to obtain a video signal, the complexity of the device settings to ensure the identity of the parameters of both CCDs, as well as the inability of the device to work with any OED (not only a CCD, but also an OED based on other types of sensors) .

An object of the invention is to increase the accuracy of automatic correction of image distortion caused by blooming.

The problem is solved in that in the known method for producing high-definition television images in a camera on conventional CCDs, including recording a video signal, automatic detection of image areas distorted by blooming OED, distortion correction by calculating the corrected pixel brightnesses of the distorted areas with subsequent recording of the corrected pixel brightnesses of the distorted image areas are introduced .

The technical problem is solved by the fact that in the device containing the ADC, the memory unit (PSU), the OED and the controller are introduced, and the OED output is connected to the first input of the ADC, the group of ADC outputs is connected to the first group of PSU inputs and the second group of controller outputs, the group of PSU outputs connected to the group of inputs of the controller, the first output of the controller is connected to the second input of the ADC, the second output of the controller is connected to the input of the PSU, the first group of outputs of the controller is connected to the second group of inputs of the PSU, the group of inputs and outputs of the controller is used to transmit corrected image.

The invention can be used to reduce image distortion caused by the presence in the field of view of the OED bright sources of radiation of small angular size, and meets the criterion of "industrial applicability".

The invention is illustrated by drawings, where figure 1 shows a structural diagram of a device, figure 2 - examples of images distorted by blooming OED, figure 3 - image after correction of distortion.

Blooming is the process of “flowing” of the charge in the OED matrix from heavily lit pixels to less lit neighboring ones, caused by overflow of a potential well in a cell (pixel). The distortions caused by blooming, appear in the form of vertical bright stripes below and above a bright object with small angular dimensions (figure 2).

The device contains (Fig. 1) OED 1, ADC 2, memory block 3, controller 4, and the output of OED 1 is connected to the first input of ADC 2, the group of outputs of ADC 2 is connected to the first group of inputs of BP 3 and the second group of outputs of controller 4, group the outputs of PSU 3 is connected to the group of inputs of controller 4, the first output of controller 4 is connected to the second input of the ADC 2, the second output of controller 4 is connected to the input of PSU 3, the first group of outputs of controller 4 is connected to the second group of inputs of PSU 3, the group of inputs and outputs of controller 4 used to transmit corrected image eniya.

The device operates as follows. The analog signal characterizing the image of the working scene, from the output of the OED 1, is fed to the first input of the ADC 2. ADC 2 performs analog-to-digital conversion and a digital value of the brightness of the current transmitted pixel is formed on the group of outputs of the ADC 2 and fed to the first group of PSU 3 inputs. From the first the output of the controller 4 to the second input of the ADC 2 receives a signal that controls the operation of the ADC 2. The second group of inputs of the BP 3 from the first group of outputs of the controller 4 receives the value of the address at which the brightness value of the current transmitted a pixel. From the second output of the controller 4, a signal is received at the input of the PSU 3, which allows information to be recorded in the memory unit 3. After the last pixel of the current frame arrives, the PSU 3 contains an image representing a matrix of pixel brightness values, possibly distorted by blooming OED.

To automatically detect image areas distorted by blooming, pixel brightness of each column of the frame is sequentially analyzed. A sign of a column belonging to a distorted image area (a sign of distortion in a column) is the simultaneous fulfillment of the following conditions:

a) a slight difference in the brightness of the pixels of the column, determined by the inequality

where Y is the vertical frame size coinciding with the vertical resolution of the OED, I _{i, j} is the brightness of the i-th pixel of the j-th analyzed column, I _{Δ pores} is the threshold value of the brightness difference, a priori set for the used OED, initially determined once in an automated mode;

(где t, m - ординаты начала и конца последовательности) пикселов столбца с яркостями I_s,j, большими или равными средней яркости

столбца, такой чтоb) the presence of a continuous sequence

(where t, m are the ordinates of the beginning and end of the sequence) of the column pixels with luminances I _{s, j} greater than or equal to the average brightness

column such that

.where I _Δtm is the threshold value of the brightness of the pixels that determines the allowable difference in the brightness of the pixels of a continuous sequence

.

The ordinates t and m of the beginning and end of the sequence are determined by the difference in brightness between adjacent pixels:

where I _ΔTM ^{is the} threshold value of the brightness of the pixels, which allows to determine the boundary of the sequence, T _M is the number of pixels sufficient to reliably determine the boundary of the sequence.

The values of I _Δtm , I _ΔTM , T _{M are} initially determined once in an automated mode for the used OED;

,

с яркостями, большими или равными средней яркости анализируемого столбца, в соседних с анализируемым столбцах. Последовательности определяются выражениямиc) the presence of continuous sequences of pixels

,

with luminances greater than or equal to the average brightness of the analyzed column, in neighboring columns. Sequences are defined by expressions

d) slight differences in the ordinates of the upper and lower boundaries of the sequences indicated in paragraphs “c” and “b”

where Δ _tm is a priori set value (as a rule, Δ _tm = 3);

пикселов столбца к максимально возможной яркости I_max пикселов кадраe) proximity to medium brightness

pixels of the column to the maximum possible brightness I _max frame pixels

where I _ΔM - a priori set value for the used OED, initially determined in an automated mode;

f) the proximity of the average brightness of the pixels of the sequences indicated in points “c” and “b” to the maximum brightness of the pixels of the frame

where I _{ΔM '} - a priori set value for the used OED, initially determined in an automated mode, as a rule I _ΔM' <I _ΔM .

When determining the distorted areas of the frame, the controller 4 at the second output generates a signal that allows reading the brightness of the pixels of the frame received at the input of the PSU 3; on the group of outputs, the controller 4 sequentially generates the addresses of the analyzed pixels arriving at the second group of inputs of the PSU 3. From the group of outputs of PSU 3 to the group of inputs of the controller 4 receives the brightness of the analyzed pixels. The controller 4 by the expressions (1) - (13) determines the columns of the frame subject to distortion by blooming.

Distortion correction is performed for separately located distorted columns and for areas of the frame containing several distorted columns located close to each other by determining the adjusted brightness of each pixel belonging to the distorted column in various ways.

каждого пиксела отдельно расположенного искаженного столбца определяется по формулеAdjusted brightness

each pixel of a separately located distorted column is determined by the formula

where I _{i, j-1} , I _{i, j + 1} are the brightness of the pixels located to the left and right of the current pixel of the adjusted column.

The correction of the brightness of the pixels of the distorted region containing two columns with abscissas j ₁ and j ₂ , and j ₁ <j ₂ , is performed according to the formulas

пиксела определяется по формулеTo correct areas containing more than two distorted columns, it is not possible to determine the corrected pixel brightnesses by interpolating by the brightnesses of neighboring pixels using expressions similar to formulas (14), (15) due to the low accuracy of interpolation and, as a consequence, the low accuracy of correction of pixel brightnesses for large areas. Correction of such areas is carried out by linear contrast, in which the adjusted brightness

pixel is determined by the formula

i ₁ > I ₁ , i ₂ <I ₂ , j ₁ > J ₁ , j ₂ <J ₂ ,

where i ₁ , i ₂ are the ordinates of the lower and upper rows included in the distorted region, j ₁ , j ₂ are the abscissas of the left and right columns included in the distorted region, I ₁ , I ₂ are the ordinates of the lower and upper rows of the region along which the parameters k, I _const , J ₁ , J ₂ are determined - the abscissas of the left and right columns of the region by which the parameters k, I _const , I ' _max , I' _min are determined - a priori set values, initially determined in an automated mode for OED.

For image correction, controller 4 calculates, according to formulas (14) - (16), the corrected pixel brightness values of distorted columns (regions) and writes the corrected values to BP 3: controller 4 sequentially generates calculated corrected pixel brightness values arriving at the first group on the second group of outputs the inputs of BP 3, the controller 4 on the first group of outputs sequentially generates the addresses received by the second group of inputs of the BP 3, by which pixel brightness is recorded, controller 4 from the second output sends to the input of BP 3 a signal that allows the recording of information in BP 3.

The sequence of operations performed by the controller 4 when correcting distortions caused by blooming is as follows: determining the distorted areas of the frame using expressions (1) - (13), correcting the brightness of the pixels of the distorted areas in accordance with expressions (14) - (16), repeating these operations until as long as no distorted column is found in determining the distorted areas of the frame.

The corrected image is transmitted for further use by external devices through the group of inputs / outputs of the controller 4.

As a controller 4, a microcomputer or a microcontroller (for example, a microcomputer described in [Technical Vision Systems: Reference Book / V.I.Syryamkin, BCTitov, Yu.G. Yakushenkov, etc.) // Under the General Editorship of V.I. Syryamkina, BC Titova, Tomsk: MGP “RASKO”, 1992. on cc.93-100 in chapter 3.6 “Microprocessors and microcomputers for vision systems”]).

The invention improves the accuracy of the correction of distortions caused by blooming OED, and provide the ability to correct distortions of any type of OED. The method allows you to automatically correct image distortion at the program level and can be used to improve the quality of photos and videos after recording (receiving).

Claims (2)

1. A method for automatically correcting blooming of an optoelectronic sensor, including recording a video signal, characterized in that it further automatically detects image areas distorted by blooming of an optoelectronic sensor (OED), corrects distortions by calculating the corrected pixel brightnesses of the distorted areas, followed by recording the corrected pixel brightnesses of the distorted areas of the image. 2. The device for automatic correction of blooming of an optoelectronic sensor containing an ADC, a memory unit (PSU), characterized in that an optoelectronic sensor (OED) and a controller are introduced into it, where the OED output is connected to the analog input of the ADC, the information group of the ADC outputs is connected to the group of data inputs of the PSU and the second group of outputs of the controller, the group of outputs of the PSU data is connected to the group of inputs of the controller, the control output of the controller is connected to the control input of the ADC, the write permission output is connected to the write permission input P, address group controller outputs coupled to address inputs of the group BP, the group controller inputs-outputs is used to transmit the corrected image.

Images