CN102857790B - Digital video static frame detection method based on time domain check - Google Patents

Abstract

The invention provides a digital video static frame detection method based on time domain check. The digital video static frame detection method mainly comprises the following steps of: setting a type and a detection threshold parameter of a signal to be detected, pre-detecting and extracting a frame synchronous signal and a frame starting limit, respectively checking the brightness Y and two chromas U and V of each row of pixel points by adopting a cyclic redundancy check (CRC) polynomial to obtain check results of each row and each frame, comparing the check result of each frame with the check result of the next frame, calculating the number of same frames, and when the number of same frames is greater than the set static frame threshold, sending a static frame indication signal and alarming. By the method, check is executed by a shift register by adopting the polynomial through simple bit XOR operation; the algorithm is simple; the calculating time is greatly shortened; the number of occupied resources is small; and the precision is high.

Description

A kind of time domain check digit digital video static frame detection

Technical field

The present invention relates to the detection method of the quiet frame of digital video image, be specifically related to a kind of based on CRC check digital video static frame detection method.

Background technology

In radio and television field of photography, in order to ensure the safety and stability of broadcast items, quiet frame detection to be carried out in multiple link to vision signal, judging whether image quiet frame and interruption occur according to testing result, to make the process such as switching in time, guarantee safety and the integrality of broadcast.

YUV is a kind of colour coding method adopted in modern color television pickup system, and wherein " Y " represents lightness, namely grey decision-making; What " U " and " V " represented is then colourity, and effect describes colors of image and saturation, is used to specify the color of pixel." brightness " is through RGB input signal to set up, and method is superimposed together by the specific part of rgb signal." colourity " then defines two aspects-tone and the saturation of color, represents respectively with Cr and Cb.At present to the detection method of the quiet frame in vision signal mainly based on to pixel Y, the weighted accumulation of U, V obtains the weighted sum of each frame, then contrasts weighted sum.Because video image vegetarian refreshments Y, U, V data volume is large, and reach the data volume of per second tens, weighted accumulation certainly will require that adder figure place is abundant and add operation is enough fast.In addition, weighted sum also needs the multiplying realizing tens with multiplier, so algorithm relative complex, takies resource many, and causes final testing result inaccuracy.

Summary of the invention

Object of the present invention will solve above-mentioned deficiency exactly, a kind of CRC check that adopts is provided to realize digital video static frame detection method, avoided complicated add operation and multiplying by position XOR and shift register, reach computing simple, take the object that resource is few, accuracy is high.

In order to achieve the above object, the invention provides a kind of time domain check digit digital video static frame detection, mainly comprise the following steps:

(1) type and the detection threshold parameter of signal to be detected is set: set pixel thresholding P, row thresholding L, frame thresholding F;

(2) treat detection signal and carry out pre-detection, obtain the type of signal to be detected;

(3) frame synchronizing signal is extracted according to the pre-detection result of (2) and frame is initial defines, and the type of output signal, obtain quiet frame threshold parameter P ', L ', F ';

(4) automatic aligning frame synchronization border, from frame boundaries, adopts CRC multinomial to verify to a pixel brightness Y and two colourity U, the V in every row the assay obtaining each row respectively:

$f(i,x\left(p\right))=\frac{x(i,p)}{\mathrm{CRC}-M}$

In formula: x represents Y, U, V of pixel respectively,

X (i, p) represents sampling Y, U, V value of p pixel of the i-th row,

CRC-M represents a M CRC multinomial, M >=16,

The value of p is 1,2,3 ..., P ',

The value of i is 1,2,3 ..., L ';

(5), after completing the pixel verification of often going, the check results of often going is counted in the check formula of this frame and carries out the cumulative check results obtaining a frame:

$\mathrm{CSx}\left(k\right)=\underset{i=1}{\overset{L^{\′}}{\mathrm{\Σ}}}\mathrm{CSx}(i,x\left(p\right))$

In formula: x represents Y, U, V of pixel respectively;

(6) check results of every frame and the check results of its next frame are compared respectively, calculate the frame number SF of same number of frames: when SF is greater than the quiet frame thresholding F ' of setting, then send quiet frame index signal, otherwise provide non-quiet frame index signal.

The preferred version of such scheme is, when the line number i+1 of present frame is not more than L ' in described step (5), then turns back in step (4) and continues to verify next line.

Wherein detection threshold parameter P ', L ', F ' contrast according to threshold parameter P, L, F of outside input the video inner parameter that corresponding signal type carries out changing.

The present invention adopts simple position XOR and shift register to complete CRC multinomial and tests, and CRC check is using number to be detected as dividend in essence, and check polynomial, as divisor, carries out without borrow division arithmetic.

Below for CRC-32, its multinomial is:

x ³²-x ²⁶+x ²³+x ²²+x ¹⁶+x ¹²+x ¹¹+x ¹⁰+x ⁸+x ⁷+x ⁵+x ⁴+x ²+x+1

The present invention has following beneficial effect compared to prior art:

1, the present invention is that employing CRC multinomial carries out simple position XOR and shift register completes inspection, and be weighted accumulative relative to traditional single frame, the algorithm of method of the present invention is relatively simple, and greatly reduces operation time;

2, the present invention utilizes FPGA, improves the compatibility of system, all has larger compatible space for different types of equipment;

3, computing circuit of the present invention is simple, so it takies, resource is few, accuracy is high.

Accompanying drawing explanation

Fig. 1 is the circuit structure diagram of CRC serial verification;

Fig. 2 is flow chart of the present invention.

Embodiment

Below in conjunction with embodiment, the present invention is further elaborated.

The register stage circuit structure chart of CRC serial verification as shown in Figure 1, before starting verification, the value of register is initialized as 1, serial signal is from the input of datin end and a high position is leading, until all data to be tested are stopped displacement by during shift-in shift register chain, namely complete verification, now the value of register is School Affairs.The verification eigenpolynomial that in figure, CRC register is corresponding is 32 ' H04C11DB7, and polynomial repressentation is as follows:

x ³²+x ²⁶+x ²³+x ²²+x ¹⁶+x ¹²+x ¹¹+x ¹⁰+x ⁸+x ⁷+x ⁵+x ⁴+x ²+x+1。

Be flow chart of the present invention as shown in Figure 2, digital video static frame detection method of the present invention comprises the steps:

(1) type and the detection threshold parameter of signal to be detected is set: set pixel thresholding P, row thresholding L, frame thresholding F;

(2) treat detection signal and carry out pre-detection, obtain the type of signal to be detected;

(3) frame synchronizing signal is extracted according to the pre-detection result of (2) and frame is initial defines, and the type of output signal, obtain quiet frame threshold parameter P ', L ', F ';

(4) automatic aligning frame synchronization border, from frame boundaries, adopts CRC multinomial to verify to a pixel brightness Y and two colourity U, the V in every row the assay obtaining each row respectively:

$f(i,x\left(p\right))=\frac{x(i,p)}{\mathrm{CRC}-M}$

In formula: x represents Y, U, V of pixel respectively;

X (i, p) represents sampling Y, U, V value of p pixel of the i-th row;

CRC-M represents a M CRC multinomial, M >=16;

The value of p is 1,2,3 ..., P ';

The value of i is 1,2,3 ..., L '.

According to verification function formula above first calculate kth frame, the i-th row Y, U, V check results as follows:

CSY(k,i)=f(i,Y(p))

CSU(k,i)=f(i,U(p))

CSV(k,i)=f(i,V(p))

Again Y, U, V of the i-th row of its next frame and k+1 frame are verified, as follows:

CSy(k+1，i)=f(i，y(p))

CSu(k+1，i)=f(i，u(p))

CSv(k+1，i)=f(i，v(p))

(5), after completing the pixel verification of often going, the check results of often going is counted in the check formula of this frame and carries out the cumulative check results obtaining a frame:

$\mathrm{CSx}\left(k\right)=\underset{i=1}{\overset{L^{\′}}{\mathrm{\Σ}}}\mathrm{CSx}(i,x\left(p\right))$

In formula: x represents Y, U, V of pixel respectively.

According to weighted formula above to each row CSy (k, i) in kth frame, CSu (k, i), CSv (k, i) carries out the cumulative integrity checking and the CSy (k) that obtain kth frame respectively, CSu (k), CSv (k).

Calculate the check results CSy (k+1, i) of each row of kth+1 frame again, CSu (k+1, i), CSv (k+1, i) cumulative integrity checking and the CSy (k+1) obtaining kth+1 frame respectively, CSu (k+1), CSv (k+1).

(6) finally by the integrity checking of kth frame and CSy (k), CSu (k), the integrity checking of CSv (k) and kth+1 frame and CSy (k+1), CSu (k+1), CSv (k+1) contrasts respectively:

When the check results of correspondence is equal, think that kth frame is identical with kth+1 frame, then the frame number SF of same number of frames now _k+1=SF _k+ 1, if SF _k+1be greater than F ' and then judge that frame is quiet frame and exports quiet frame indication warning for this reason; Otherwise be judged as normal video frame, export non-quiet frame instruction, now K+1 frame being set as new reference frame, is SF _k+1=0 re-starts detection.

Only that the preferred embodiment of the present invention is described above; not scope of the present invention is limited; do not departing under the present invention designs spiritual prerequisite; the various distortion that the common engineers and technicians in this area make technical solution of the present invention and improvement, all should fall in protection range that claims of the present invention determine.

Claims (2)

1. a time domain check digit digital video static frame detection, is characterized in that: comprise the following steps:

(1) type and the detection threshold parameter of signal to be detected is set: set pixel thresholding P, row thresholding L, frame thresholding F;

(2) treat detection signal and carry out pre-detection, obtain the type of signal to be detected;

(3) frame synchronizing signal is extracted according to the pre-detection result of (2) and frame is initial defines, and the type of output signal, obtain quiet frame threshold parameter P ', L ', F ';

(4) automatic aligning frame synchronization border, from frame boundaries, adopts CRC multinomial to verify to brightness Y, the colourity U of every row pixel, V the assay obtaining each row respectively:

In formula: x represents Y, U, V of pixel respectively,

X (i, p) represents sampling Y, U, V value of p pixel of the i-th row,

CRC-M represents a M CRC multinomial, M >=16,

The value of p is 1,2,3 ..., P ',

The value of i is 1,2,3 ..., L ';

Again according to above-mentioned assay calculate kth frame, the i-th row Y, U, V check results as follows:

CSY(k,i) = f( i,Y(p) )

CSU(k,i)= f( i, U(p) )

CSV(k,i)=f( i, V(p) )；

(5), after completing the pixel verification of often going, the check results of often going is counted in the check formula of this frame and carries out the cumulative check results obtaining a frame:

In formula: x represents Y, U, V of pixel respectively;

(6) check results of every frame and the check results of its next frame are compared respectively, when the check results of correspondence is equal, think that this two frame is identical, then draw the frame number SF calculating same number of frames: when SF is greater than the quiet frame thresholding F ' of setting, then send quiet frame index signal, otherwise provide non-quiet frame index signal.

2. a kind of time domain check digit digital video static frame detection according to claim 1, is characterized in that: when in described step (5), the line number i+1 of present frame is not more than L ', then turn back in step (4) and continue next line verification.