CN105847708A - Image-histogram-analysis-based automatic exposure adjusting method and system for linear array camera - Google Patents

Abstract

The invention provides an image-histogram-analysis-based automatic exposure adjusting method and system for a linear array camera. The method comprises: a grey-level histogram of a current image is calculated, normalization is carried out on the grey-level histogram, and a left boundary, a right boundary and a peak value of the histogram of the current image are obtained; according to an expected peak value of a grey-level histogram of an image, an exposure adjusting value of a camera is calculated; if an absolute value of the exposure adjusting value is larger than or equal to an adjusting threshold value, a new current exposure value is calculated; if the absolute value is less than the adjusting threshold value, the current exposure value is kept without any adjustment; shooting is carried out based on the current exposure value; and after a new image is obtained and is used as a current image, adjustment is carried out continuously. According to the invention, a histogram analysis is carried out on obtained image data and exposure time adjustment of the camera is carried out by using the expected image quality and the histogram distribution feature, so that an effect of automatic exposure time adjustment of the camera is realized.

Description

Line-scan digital camera automatic exposure method of adjustment based on image histogram analysis and system

Technical field

The present invention relates to digital image processing field, particularly relate to a kind of line-scan digital camera based on image histogram analysis automatic Exposure adjustment method and system.

Background technology

To country sport target (such as train) imaging, it is contemplated that move and splicing factor, many employing line-scan digital cameras are taken pictures. For the environment of field image data acquiring, the light filling of active light source is the necessary condition taken pictures, but external environment condition Change, such as the interference of the phenomenons such as night, fine day, overcast and rainy, direct sunlight, motion, how to effectively eliminate, obtain Obtain better image effect, be the problem that must solve.The histogram analysis of image is generally used for the statistical analysis of image, Can be used for equalization and (such as " histogram equalization method of a kind of infrared image " (CN 104252700A), but rarely have use In the method reversely instructing regulation camera exposure parameter.

By the means such as photometry apparatus, luminous intensity measurement, the change to external ambient light measures, and needs to consider shadow region The factors such as territory, external light source interference, object of which movement, and to the phenomenon such as imaging object self is reflective, the most do not compare Good solution.Therefore, it is necessary to provide a kind of new camera exposure method automatically to adjust, to facilitate acquisition light and shade Suitably, imaging effect dynamic scene picture clearly.

Summary of the invention

In order to solve the problems referred to above, the present invention proposes a kind of line-scan digital camera automatic exposure based on image histogram analysis and adjusts Method and system, from image quality is analyzed, are embodied in external interference factor in imaging results, pass through image histogram Analyze to obtain and be suitable for current environment and the time for exposure of current time, thus obtain high quality graphic.

Technical scheme provides a kind of line-scan digital camera automatic exposure method of adjustment based on image histogram analysis, its It is characterised by, comprises the following steps:

Step 1, acquisition present image, calculate the grey level histogram of present image, be normalized grey level histogram, Obtain histogrammic left margin Left_Pos, right margin Rigt_Pos and the peak value of present image；

Step 2, expectation peak E xp_Bright according to image grey level histogram, calculate camera exposure adjusted value CExp It is as follows,

$\begin{array}{c}CExp=\\ \left\{\begin{array}{cc}Exp\_Bright-\frac{({\mathrm{Left}}_{Pos}+{\mathrm{Right}}_{Pos})}{2}\×k1\÷k2+0.5& Left\_Pos\≠0,Right\_Pos\≠255\\ Exp\_Bright-\frac{(Left\_Sec\_Pos+Right\_Sec\_Pos)}{2}\×k1\÷k2+0.5& Left\_Pos=0\\ Exp\_Bright-\frac{(Left\_Sec\_Pos+Right\_Sec\_Pos)}{2}\×k1\÷k2+0.5& Right\_Pos=255\end{array}\right.\end{array}$

Wherein, Exp_Bright is the expectation peak value of image histogram, Left_Sec_Pos and Rigt_Sec_Pos is for rejecting border Secondary boundary value after peak value, correction factor k1, k2 are image histogram linear transform coefficient；

Step 3, the absolute value comparing camera exposure adjusted value CExp and default adjustment threshold value, if camera exposure is adjusted Camera exposure adjusted value and current exposure value, more than or equal to adjusting threshold value, are added by the absolute value of whole value CExp, obtain new Current exposure value, enter step 4；If the absolute value of camera exposure adjusted value CExp is less than adjusting threshold value, the most not It is adjusted, keeps current exposure value, enter step 4；

Step 4, take pictures according to current exposure value, when obtain new image as present image after, return to step Rapid 1, continue to adjust.

And, according to default filtering threshold, histogram is intercepted, update the histogrammic left margin of present image Left_Pos, right margin Rigt_Pos and peak value.

And, the line number of present image uses the numerical value m preset.

And, m=1000.

And, the numerical value n, n according to default > 1, when step 4 is taken pictures according to current exposure value, it is to wait for clapping Just extract up-to-date m row image after taking the photograph (n-1) × m row as new present image, then return to step 1, Continue to adjust.

The present invention correspondingly provide a kind of line-scan digital camera automatic exposure based on image histogram analysis adjust system, including with Lower module:

First module, is used for obtaining present image, calculates the grey level histogram of present image, returns grey level histogram One changes, and obtains histogrammic left margin Left_Pos, right margin Rigt_Pos and the peak value of present image；

Second module, for expectation peak E xp_Bright according to image grey level histogram, calculates camera exposure adjusted value CExp is as follows,

$\begin{array}{c}CExp=\\ \left\{\begin{array}{cc}Exp\_Bright-\frac{({\mathrm{Left}}_{Pos}+{\mathrm{Right}}_{Pos})}{2}\×k1\÷k2+0.5& Left\_Pos\≠0,Right\_Pos\≠255\\ Exp\_Bright-\frac{(Left\_Sec\_Pos+Right\_Sec\_Pos)}{2}\×k1\÷k2+0.5& Left\_Pos=0\\ Exp\_Bright-\frac{(Left\_Sec\_Pos+Right\_Sec\_Pos)}{2}\×k1\÷k2+0.5& Right\_Pos=255\end{array}\right.\end{array}$

Wherein, Exp_Bright is the expectation peak value of image histogram, Left_Sec_Pos and Rigt_Sec_Pos is for rejecting border Secondary boundary value after peak value, correction factor k1, k2 are image histogram linear transform coefficient；

3rd module, for the absolute value comparing camera exposure adjusted value CExp and the adjustment threshold value preset, if camera Camera exposure adjusted value and current exposure value, more than or equal to adjusting threshold value, are added by the absolute value of exposure adjusted value CExp, Obtain new current exposure value, order the 4th module work；If the absolute value of camera exposure adjusted value CExp is less than adjusting Whole threshold value, is not adjusted, and keeps current exposure value, order the 4th module work；

4th module, for taking pictures according to current exposure value, when obtain new image as present image after, order First module work, continues to adjust.

And, according to default filtering threshold, histogram is intercepted, update the histogrammic left margin of present image Left_Pos, right margin Rigt_Pos and peak value.

And, the line number of present image uses the numerical value m preset.

And, m=1000.

And, the numerical value n, n according to default > 1, when the 4th module is taken pictures according to current exposure value, it is to wait for Just extract up-to-date m row image as new present image after shooting (n-1) × m row, then order the first module Work, continues to adjust.

The present invention, by the view data obtained is carried out histogram analysis, utilizes variations in peak and histogram waveform special The change levied, the structure time for exposure is with the corresponding relation of peak change, thus utilizes the desired quality of image and its histogram The camera exposure time is adjusted by distribution characteristics, makes the histogram feature of the view data of acquisition match with desired value, thus Reach camera and automatically adjust the effect of time for exposure.

Accompanying drawing explanation

Fig. 1 is the application system system assumption diagram of the embodiment of the present invention；

Fig. 2 is the front image histogram obtained of embodiment of the present invention regulation.

Fig. 3 be the embodiment of the present invention regulation after obtain correspondence image histogram.

Specific implementation method

Understand and implement the present invention for the ease of those of ordinary skill in the art, below in conjunction with the accompanying drawings and detailed description of the invention pair The present invention is described in further detail.

The embodiment of the present invention is provided a kind of line-scan digital camera automatic exposure method of adjustment based on image histogram analysis, including Following steps:

Step 1, acquisition present image, calculate the grey level histogram of present image, be normalized grey level histogram, Obtain histogrammic left margin, right margin and the peak value of present image；

The grey level histogram of present image in step 1, is normalized by embodiment, disturbs for attenuating noise, Embodiment arranges histogrammic filtering threshold k=0.3, histogram is intercepted, thus obtain the Nogata of present image Scheme new left margin Left_Pos, right margin Rigt_Pos and peak value.When being embodied as, those skilled in the art can be voluntarily Presetting filtering threshold, employing empirical value can be with the impact of attenuating noise.

Step 2, expectation peak value according to image grey level histogram, calculate camera exposure adjusted value；

In step 2, if left margin Left_Pos=0 or right margin Rigt_Pos=255, then to image on a new left side Seek the cumulative distribution of image between right peak value, thus realize histogram equalization, get the newest histogrammic left margin Left_Sec_Pos, right margin Rigt_Sec_Pos.

In embodiment, input ideal image histogrammic expectation peak parameters Exp_Bright, calculate camera according to formula and expose Light adjusted value CExp:

$\begin{array}{c}CExp=\\ \left\{\begin{array}{cc}Exp\_Bright-\frac{({\mathrm{Left}}_{Pos}+{\mathrm{Right}}_{Pos})}{2}\×k1\÷k2+0.5& Left\_Pos\≠0,Right\_Pos\≠255\\ Exp\_Bright-\frac{(Left\_Sec\_Pos+Right\_Sec\_Pos)}{2}\×k1\÷k2+0.5& Left\_Pos=0\\ Exp\_Bright-\frac{(Left\_Sec\_Pos+Right\_Sec\_Pos)}{2}\×k1\÷k2+0.5& Right\_Pos=255\end{array}\right.\end{array}$

Wherein, Exp_Bright is the expectation peak value of image histogram, Left_Sec_Pos and Rigt_Sec_Pos is for rejecting border Secondary boundary value after peak value.Wherein correction factor k1, k2 is image histogram linear transform coefficient, during specific embodiment Those skilled in the art can preset value voluntarily, such as, can obtain by changing the time for exposure of camera in fixed environment Standard picture gray average statistics obtains, and embodiment takes k1=5, k2=6.

Step 3, for ensure system call interception efficiency, when being embodied as, those skilled in the art can preset voluntarily adjust threshold Value, the biggest efficiency of value is the highest, adjusts accuracy the lowest, and the least regulated efficiency of value is the lowest, and Adjustment precision is the highest.Real Executing example and arranging adjustment threshold value is 10, i.e. camera exposure adjusted value absolute value is more than or equal to 10, then explanation effect is the most undesirable, Camera exposure adjusted value and current exposure value are added, obtain new current exposure value, enter step by new current exposure value Rapid 4, take pictures, when obtain new image as present image after, step 1, repeated execution of steps 1, step will be returned Rapid 2 and step 3 continue adjust；If exposure adjusted value absolute value is less than 10, then currently without adjusting, camera is made to adjust Value, equal to zero, keeps using current exposure value, enters step 4；

Step 4, according to current exposure value, camera is normally taken pictures, obtain image.

In step 4, take pictures according to Current camera exposure value, continue take new image as present image after, Return to step 1, continue to adjust, to reach persistently to obtain the purpose of picture rich in detail.Until halt system works.

The system architecture citing of application technical solution of the present invention is as it is shown in figure 1, after initial camera captures image, hold The computer of the above flow process of row can monitor image, adjusts camera parameter, and stores server, and server exports The corresponding command, to industrial computer, controls to adjust the exposure value of line-scan digital camera.Such as, can be used for the fire of outdoor dynamic operation Car is monitored, and obtains picture rich in detail, in order to follow-up vehicle monitoring, identification.So, main use environment is family External environment, use condition should adapt to multiple Changes in weather.When being embodied as, 1 group of inverted "L" shaped support can be used； The cross bar of inverted "L" shaped support is installed on industrial camera shooting top image, a montant taken pictures vertically downward One industrial camera being perpendicular to train running direction is installed, and increases some light filling equipment, such as light source led. Hardware environment can be carried out the collecting work of image after putting in place right, uses technical solution of the present invention to obtain the first width After image, regulation, according to external environment condition, is automatically adjusted the time for exposure, obtains ideal image.

According to the characteristic of line-scan digital camera, when being embodied as, those skilled in the art can pre-set image computing unit voluntarily, The i.e. line number of present image uses the numerical value m preset.The present embodiment takes 1000 initial row images as an image meter Calculate unit, obtain its histogram information (such as Fig. 2), in conjunction with before time for exposure, obtain time for exposure adjusted value (such as Fig. 3), is then adjusted the exposure value of camera, is to obtain 1000 row data after adjusting equally.I.e. when Front image always takes m=1000 row view data.

During adjusting, camera normal acquisition image, in order to avoid adjusting exposure value the most frequently, therefore Present invention further propose that, no matter by new current exposure value or retaining original current exposure value takes pictures, can Not being to obtain new image as present image at once, and just extract up-to-date m after being to wait for shooting (n-1) × m row Row image is as new present image.In embodiment, close to for every n × 1000 Row sum-equal matrix once in image appearance Time for exposure, wherein regulated efficiency as required can set n value, n > 1, when being embodied as, people in the art Member can preset the value of n voluntarily.N=2 in embodiment, until adjusting end-of-job.

When being embodied as, method provided by the present invention can realize automatic operational process based on software engineering, it is possible to uses module Change mode realizes corresponding system.

The embodiment of the present invention correspondingly provides a kind of line-scan digital camera automatic exposure based on image histogram analysis and adjusts system, Including with lower module:

First module, is used for obtaining present image, calculates the grey level histogram of present image, returns grey level histogram One changes, and obtains histogrammic left margin Left_Pos, right margin Rigt_Pos and the peak value of present image；

Second module, for expectation peak E xp_Bright according to image grey level histogram, calculates camera exposure adjusted value CExp is as follows,

$\begin{array}{c}CExp=\\ \left\{\begin{array}{cc}Exp\_Bright-\frac{({\mathrm{Left}}_{Pos}+{\mathrm{Right}}_{Pos})}{2}\×k1\÷k2+0.5& Left\_Pos\≠0,Right\_Pos\≠255\\ Exp\_Bright-\frac{(Left\_Sec\_Pos+Right\_Sec\_Pos)}{2}\×k1\÷k2+0.5& Left\_Pos=0\\ Exp\_Bright-\frac{(Left\_Sec\_Pos+Right\_Sec\_Pos)}{2}\×k1\÷k2+0.5& Right\_Pos=255\end{array}\right.\end{array}$

Wherein, Exp_Bright is the expectation peak value of image histogram, Left_Sec_Pos and Rigt_Sec_Pos is for rejecting border Secondary boundary value after peak value, correction factor k1, k2 are image histogram linear transform coefficient；

3rd module, for the absolute value comparing camera exposure adjusted value CExp and the adjustment threshold value preset, if camera Camera exposure adjusted value and current exposure value, more than or equal to adjusting threshold value, are added by the absolute value of exposure adjusted value CExp, Obtain new current exposure value, order the 4th module work；If the absolute value of camera exposure adjusted value CExp is less than adjusting Whole threshold value, is not adjusted, and keeps current exposure value, order the 4th module work；

4th module, for taking pictures according to current exposure value, when obtain new image as present image after, order First module work, continues to adjust.

Each module implements and can be found in corresponding steps, and it will not go into details for the present invention.

The above is only to illustrate that the present invention's is embodied as case, is not limited to implementing of the present invention Scope, such as those skilled in the art completed under without departing from the spirit indicated by the present invention and principle all etc. Effect changes or modifies, and must be covered by the scope of the claims in the present invention.

Claims (10)

1. a line-scan digital camera automatic exposure method of adjustment based on image histogram analysis, it is characterised in that comprise the following steps:

Step 1, acquisition present image, calculate the grey level histogram of present image, be normalized grey level histogram, Obtain histogrammic left margin Left_Pos, right margin Rigt_Pos and the peak value of present image；

Step 2, expectation peak E xp_Bright according to image grey level histogram, calculate camera exposure adjusted value CExp It is as follows,

$\begin{array}{c}CExp=\\ \left\{\begin{array}{cc}Exp\_Bright-\frac{({\mathrm{Left}}_{Pos}+{\mathrm{Right}}_{Pos})}{2}\×k1\÷k2+0.5& Left\_Pos\≠0,Right\_Pos\≠255\\ Exp\_Bright-\frac{(Left\_Sec\_Pos+Right\_Sec\_Pos)}{2}\×k1\÷k2+0.5& Left\_Pos=0\\ Exp\_Bright-\frac{(Left\_Sec\_Pos+Right\_Sec\_Pos)}{2}\×k1\÷k2+0.5& Right\_Pos=255\end{array}\right.\end{array}$

Wherein, Exp_Bright is the expectation peak value of image histogram, Left_Sec_Pos and Rigt_Sec_Pos is for rejecting border Secondary boundary value after peak value, correction factor k1, k2 are image histogram linear transform coefficient；

Step 3, the absolute value comparing camera exposure adjusted value CExp and default adjustment threshold value, if camera exposure is adjusted Camera exposure adjusted value and current exposure value, more than or equal to adjusting threshold value, are added by the absolute value of whole value CExp, obtain new Current exposure value, enter step 4；If the absolute value of camera exposure adjusted value CExp is less than adjusting threshold value, the most not It is adjusted, keeps current exposure value, enter step 4；

Step 4, take pictures according to current exposure value, when obtain new image as present image after, return to step Rapid 1, continue to adjust.

Line-scan digital camera automatic exposure method of adjustment based on image histogram analysis the most according to claim 1, it is characterised in that: According to default filtering threshold, histogram is intercepted, update present image histogrammic left margin Left_Pos, Right margin Rigt_Pos and peak value.

Line-scan digital camera automatic exposure method of adjustment based on image histogram analysis the most according to claim 1 or claim 2, its feature It is: the line number of present image uses the numerical value m preset.

Line-scan digital camera automatic exposure method of adjustment based on image histogram analysis the most according to claim 3, it is characterised in that: M=1000.

Line-scan digital camera automatic exposure method of adjustment based on image histogram analysis the most according to claim 3, it is characterised in that: Numerical value n, n according to default > 1, when taking pictures according to current exposure value in step 4, is to wait for shooting (n-1) × m Just extract up-to-date m row image after row as new present image, then return to step 1, continue to adjust.

6. a line-scan digital camera automatic exposure based on image histogram analysis adjusts system, it is characterised in that include with lower module:

First module, is used for obtaining present image, calculates the grey level histogram of present image, carries out grey level histogram Normalization, obtains histogrammic left margin Left_Pos, right margin Rigt_Pos and the peak value of present image；

Second module, for expectation peak E xp_Bright according to image grey level histogram, calculates camera exposure and adjusts CExp is as follows for value,

$\begin{array}{c}CExp=\\ \left\{\begin{array}{cc}Exp\_Bright-\frac{({\mathrm{Left}}_{Pos}+{\mathrm{Right}}_{Pos})}{2}\×k1\÷k2+0.5& Left\_Pos\≠0,Right\_Pos\≠255\\ Exp\_Bright-\frac{(Left\_Sec\_Pos+Right\_Sec\_Pos)}{2}\×k1\÷k2+0.5& Left\_Pos=0\\ Exp\_Bright-\frac{(Left\_Sec\_Pos+Right\_Sec\_Pos)}{2}\×k1\÷k2+0.5& Right\_Pos=255\end{array}\right.\end{array}$

Wherein, Exp_Bright is the expectation peak value of image histogram, Left_Sec_Pos and Rigt_Sec_Pos is for rejecting border Secondary boundary value after peak value, correction factor k1, k2 are image histogram linear transform coefficient；

3rd module, for the absolute value comparing camera exposure adjusted value CExp and the adjustment threshold value preset, if camera Camera exposure adjusted value and current exposure value, more than or equal to adjusting threshold value, are added by the absolute value of exposure adjusted value CExp, Obtain new current exposure value, order the 4th module work；If the absolute value of camera exposure adjusted value CExp is less than adjusting Whole threshold value, is not adjusted, and keeps current exposure value, order the 4th module work；

4th module, for taking pictures according to current exposure value, when obtain new image as present image after, life Make the first module work, continue to adjust.

The most according to claim 6, line-scan digital camera automatic exposure based on image histogram analysis adjusts system, it is characterised in that: According to default filtering threshold, histogram is intercepted, update present image histogrammic left margin Left_Pos, Right margin Rigt_Pos and peak value.

8. adjust system, its feature according to line-scan digital camera automatic exposure based on image histogram analysis described in claim 6 or 7 It is: the line number of present image uses the numerical value m preset.

The most according to claim 8, line-scan digital camera automatic exposure based on image histogram analysis adjusts system, it is characterised in that: M=1000.

The most according to claim 8, line-scan digital camera automatic exposure based on image histogram analysis adjusts system, and its feature exists In: the numerical value n, n according to default > 1, when the 4th module is taken pictures according to current exposure value, it is to wait for shooting (n- 1) just extract up-to-date m row image after × m row as new present image, then order the first module work, continue Continuous adjustment.