JP4414289B2 - Contrast enhancement imaging device - Google Patents

Description

  The present invention relates to a contrast-enhanced imaging apparatus that corrects an image to an appropriate brightness when the image captured by a camera or the like is underexposed or overexposed.

  When a conventional contrast-enhanced image pickup device captures an image using the image pickup device, in order to prevent overexposure and underexposure in an image, the degree of highlight is determined based on luminance information at one location in the shooting screen division area, A gamma correction is performed by selecting an optimal curve from a plurality of registered gamma curves (see, for example, Patent Document 1), or a gamma correction that raises the luminance level of a low-brightness region based on automatic exposure control and exposure amount. A thing (for example, refer patent document 2) is known.

Patent No. 2935116 Japanese Patent Laid-Open No. 2004-23605

  However, when a conventional contrast-enhanced imaging device captures an image having a plurality of high-luminance subjects and low-luminance subjects using the imaging device, the imaging device described in the former Patent Document 1 divides the imaging screen. However, since only correction based on luminance information at one location is performed, the illuminance of the subject is reduced when the subject is imaged in an apparatus that needs to shoot an image outdoors, such as a vehicle number recognition device or an entrance / exit monitoring device. If it is insufficient or excessive, or if there is uneven brightness in one image, it is not possible to select the optimal gamma curve for each subject with different brightness, and the exact vehicle number, etc. is recognized. It was difficult to do.

  An object of the present invention is to provide a contrast-enhanced imaging apparatus capable of obtaining a good image by performing luminance correction on each of a plurality of subjects having different luminances.

  In order to achieve the above object, the present invention provides a video input unit that stores video from the imaging unit, a correction amount calculation unit that calculates a contrast correction amount from video luminance information of the video input unit, and a correction amount calculation unit In the contrast-enhanced imaging device including a video correction unit that corrects a video image using the correction amount calculated in step S1 and a video output unit that outputs a video image corrected by the video correction unit, the correction amount calculation unit includes the video input unit. A data reading unit for reading a predetermined amount of data from the data, a local maximum point detecting unit and a local minimum point detecting unit for detecting the local maximum luminance value and the local minimum luminance value from the predetermined amount of data, and the predetermined amount of data divided into a plurality of blocks The intermediate point calculation unit for calculating the intermediate point luminance value from the maximum point luminance value and the minimum point luminance value in each of the divided blocks, and the maximum point luminance value and the minimum point luminance value in each block. A bandwidth calculation unit that calculates a bandwidth; and a coefficient calculation unit that calculates a contrast enhancement coefficient for contrast enhancement from each of the intermediate point luminance values and the respective bandwidths, and the video correction unit includes: The image correction is performed for each block based on the contrast enhancement coefficient.

  According to a second aspect of the present invention, in the first aspect, the data reading unit reads at least one line data from the video input unit, and the intermediate point calculation unit is a pixel of the one line data. The intermediate point luminance value is calculated from the maximum point luminance value and the minimum point luminance value in the block which is divided by having one maximum point luminance value and one minimum point luminance value in the direction.

  Further, the present invention described in claim 3 is the one described in claim 1, wherein the data reading unit sequentially reads out one line data from the video input unit, and the intermediate point calculating unit reads out the one line data. And calculating a midpoint luminance value from a maximum point luminance value and a minimum point luminance value in a block having one maximum point luminance value and one minimum point luminance value in the pixel direction.

  Furthermore, the present invention according to claim 4 is the one according to claim 1, wherein the intermediate point calculation unit is configured such that a difference between adjacent maximum point luminance values and minimum point luminance values in the block is a predetermined first set value. The present invention is characterized in that the intermediate point luminance value is not output in the following cases.

  According to the contrast-enhanced imaging device of the present invention, since contrast enhancement correction is performed on the basis of the midpoint luminance value and the bandwidth for each of the blocks divided into a plurality, the weather, sunlight Even if the illuminance of the subject is insufficient or excessive due to the orientation of the image, or even if there is uneven brightness in one image, the contrast enhancement correction can be performed on the image in real time in the imaging unit, which is stable A contrast image can be obtained.

  According to a second aspect of the present invention, there is provided the contrast-enhanced imaging device based on the intermediate point luminance value and the bandwidth for each block divided into the maximum point luminance value and the minimum point luminance value in the pixel direction of at least one line. Since contrast enhancement correction is performed, for example, when an image of a subject is specified in the line direction, the contrast can be adjusted appropriately along the line direction of the specific region, and stable. Contrast images can be obtained.

  Furthermore, the contrast-enhanced imaging device according to the present invention described in claim 3 is provided with an intermediate-point luminance value and a band for each block divided for each line data and for each block divided into a maximum point luminance value and a minimum point luminance value in the pixel direction. Since contrast enhancement correction is performed on the basis of the width, a stable contrast image can be obtained over the entire region regardless of where the subject from which the detailed data is obtained is shown.

  Furthermore, the contrast-enhanced imaging device according to the present invention described in claim 4 is configured not to output the intermediate point luminance value when the difference between the adjacent maximum point luminance value and the minimum point luminance value in the block is equal to or smaller than a predetermined value. The contrast is not overly corrected by noise mixed in the block, and a stable contrast image can be obtained.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 is a block diagram showing a contrast-enhanced imaging device according to an embodiment of the present invention.
The contrast-enhanced imaging device includes a video input unit 11 that stores a video from the imaging unit 10 including a lens 10a and an imaging element 10b, and a correction amount that calculates a correction amount for video correction from video luminance information of the video input unit 11. A calculation unit 12, a video correction unit 13 that corrects video with the correction amount calculated by the correction amount calculation unit 12, and a video output unit 14 that outputs a corrected video from the video correction unit 13 are provided. The correction amount calculation unit 12 described above includes a data reading unit 12a that reads a predetermined amount, for example, data for each line, from the video input unit 11, a maximum point detection unit 12b that detects a maximum point luminance value from the predetermined amount of data, A local minimum point detection unit 12c that detects a local minimum point luminance value from a predetermined amount of data, and a local maximum point luminance value and a local minimum point luminance value that are adjacent to each other by the local maximum point detection unit 12b and the local minimum point detection unit 12c in a block that will be described in detail later. An intermediate point calculation unit 12d for calculating an intermediate point luminance value, a bandwidth calculation unit 12e for calculating a bandwidth from adjacent maximum point luminance values and minimum point luminance values, and an intermediate point luminance value calculated by the intermediate point calculation unit 12d And a coefficient calculation unit 12f for calculating a contrast enhancement coefficient for contrast enhancement from the bandwidth, and a coefficient holding unit 12g for holding the contrast enhancement coefficient.

FIG. 2 is a flowchart showing the operation of the above-described contrast-enhanced imaging device.
For example, when an image sensor of horizontal 1280 pixels × vertical 1024 pixels is used as the image sensor 10b of the imaging unit 10 that captures an image of a subject, one line becomes 1280 luminance data, and a total of 1024 lines of the data is in step S1. Input to the video input unit 11. Subsequently, the correction amount calculation unit 12 calculates a correction amount for contrast enhancement correction. When the luminance data for a predetermined amount, here, one line is read by the data reading unit 12a in step S2, waveform diagram data as shown in FIG. 4 is obtained. In step S3, the maximum point luminance value that shifts from rising to falling in one line data is detected by the maximum point detecting unit 12b, and in step S4, the minimum point luminance value that shifts from falling to rising is detected by the minimum point detecting unit 12c. In step S <b> 5, the midpoint detection unit 12 d calculates a midpoint luminance value that is located between the adjacent maximum point luminance value and the minimum point luminance value. In addition, the bandwidth calculation unit 12e calculates the difference between the maximum point luminance value obtained from the midpoint luminance value and the minimum point luminance value as the bandwidth in step S6.

  At this time, the beginning and end of the data in one line data are set as a block 1 and a block 12 so as to include one maximum point luminance value and one minimum point luminance value, and the adjacent maximum point luminance values and minimum values are between them. The point luminance values are designated as block 2 to block 11, respectively. The maximum point luminance values existing in the respective blocks 1 to 12 are indicated by MAX1 to MAX7, and the minimum point luminance values are indicated by MIN1 to MIN6. Here, there is a change point indicated as point J in the block 10, but the midpoint detection unit 12d has a difference between the adjacent maximum point luminance value and the minimum point luminance value in the block equal to or less than a predetermined first set value. In some cases, the midpoint luminance value is not calculated and is not output as a result, thereby preventing sensitive contrast correction due to noise mixing.

  Subsequently, in step S7, the coefficient calculation unit 12f calculates a gamma coefficient for enhancing contrast based on the calculated midpoint luminance value and bandwidth, respectively, and obtains a contrast enhancement correction coefficient. When the video capture interval is high, for example, 60 fps (frame / second), the video correction unit 13 corrects one line data with the contrast enhancement correction coefficient in step S8, and outputs it from the video output unit 14 in step S9. To do. In step S10, it is determined whether or not the above-described processing has been output for all lines, and processing is repeated until output is performed for all lines, thereby generating an image with contrast in real time.

  Further, when the video capture interval is low, for example, 15 fps and real time is not required for video capture, after calculating the contrast enhancement correction coefficient for enhancing the contrast in step S7 as shown in the flowchart of FIG. In step 11, the coefficient is held in the coefficient holding unit 12g. In step S12, it is determined whether or not the processing described above has been performed for all lines, and the processing is repeated until all lines are processed. Thereafter, using the above-described contrast enhancement correction coefficients for all lines held in step 13, the video correction unit 13 may perform video correction for all lines at once, and may output from the video output unit 14 in step 14.

  When the bandwidth calculated for a certain block in the one-line data described above is smaller than the predetermined second setting value set in advance, the contrast in the block is insufficient, and the block is emphasized, and further in the middle. When the point luminance value is lower than the predetermined luminance value, as shown in the comparison with FIG. 7 which is a correction curve when the gamma coefficient is 1, an intermediate value like the contrast enhancement correction characteristic curve shown in FIG. When the point luminance value is a coefficient-delimited luminance point k1, an increase in the gamma coefficient β1 that brightens the luminance is corrected largely, and a decrease in the gamma coefficient α1 that darkens the luminance is corrected small.

  On the other hand, when the midpoint luminance value is higher than the predetermined second set value in the block that emphasizes contrast because the contrast is insufficient, it can be seen from comparison with FIG. 7 that is a correction curve when the gamma coefficient is 1. As described above, when the intermediate point luminance value is a coefficient-separated luminance point k2 as in the characteristic curve for contrast enhancement correction shown in FIG. 6, the increase in the gamma coefficient β2 that increases the luminance is reduced, and the gamma coefficient α2 that decreases the luminance. Correct the decrease of. However, the gamma coefficients α and β are based on 1.

  In the above-described contrast enhancement correction, conversion is performed according to Equation 1, where the luminance before conversion is src, the luminance after conversion is dst, the gamma coefficient for decreasing the luminance is α, the gamma coefficient for increasing the luminance is β, and the coefficient separation luminance is k. .

(Equation 1)
dst = ((src / k) ^ (1 / α)) × k
However, 0 <k <255, 0 ≦ src ≦ k, and 0 <α ≦ 1.

(Equation 2)
dst = ((src / (255−k)) ^ (1 / β)) × (255−k)
However, 0 <k <255, k <src ≦ 255, 1 ≦ β, and ^ indicate powers.

  In this way, contrast enhancement correction is performed for each block based on the midpoint luminance value and the bandwidth, so that it is necessary for devices that need to shoot images outdoors, such as vehicle number recognition devices and entrance / exit monitoring devices. If it is used, the contrast enhancement correction will be applied to the image even if the subject's illumination is insufficient or excessive due to the weather, the direction of sunlight, etc., or when there is uneven brightness in one image when imaging the subject. This can be performed in real time in the imaging unit, and a stable contrast image can be obtained to accurately recognize the vehicle number and the like. According to a preferred embodiment, a plurality of subjects with different luminances are divided into blocks, and luminance correction is performed on the respective blocks to obtain a good image with contrast in each of the high, medium, and low luminance regions. Can be obtained.

  Further, since the midpoint detection unit 12d is configured not to calculate or output the midpoint luminance value when the difference between the adjacent maximum point luminance value and the minimum point luminance value in the block is equal to or less than a predetermined first set value, Therefore, the contrast is not corrected excessively by noise mixed in the image, and a stable contrast image can be obtained.

  In the embodiment described above, one line data of video data is handled as a unit as shown in FIG. 4, but a plurality of lines or all lines can be handled as a unit. The pixel direction has one maximum point luminance value and one minimum point luminance value and is treated as one block. However, all image data is divided into a plurality of blocks according to the number of pixels, as shown in FIG. A plurality of blocks are collectively divided into one block, and for each of the divided blocks, an average midpoint luminance value is calculated from an average value of each local maximum luminance value and an average value of each local minimum luminance value, Brightness correction can also be performed. Also, the maximum point luminance value and minimum point luminance value of the previous block and the maximum point luminance value and minimum point luminance value of the next block, or the intermediate point luminance value of the previous block and the intermediate point luminance value of the next block If the change is within the range of the predetermined third set value, the midpoint luminance value and the bandwidth of the previous block can be applied to the next block and simplified.

  As described above, at the time of carrying out the present invention, at least one line data among all the image data may be divided into a plurality of blocks in the pixel direction. However, as shown in FIG. In the pixel direction, the block divided into the maximum point luminance value and the minimum point luminance value is finer, and desirable luminance correction can be performed according to each position.

  When such a contrast-enhanced imaging device is used as a device that needs to shoot images outdoors, such as a vehicle number recognition device or an entrance / exit monitoring device, the illuminance of the subject is insufficient or excessive when the subject is imaged. In this case, even when there is uneven brightness in one image, it is possible to perform contrast enhancement correction on the image in real time in the imaging unit and obtain an image with stable contrast.

  The contrast-enhanced imaging apparatus according to the present invention is not limited to the above-described embodiment, and can be applied to other contrast-enhanced imaging apparatuses.

It is a block block diagram which shows the contrast emphasis imaging device by one embodiment of this invention. 3 is a flowchart illustrating an operation of a main part in the contrast-enhanced imaging device illustrated in FIG. 1. 6 is a flowchart showing another operation of the main part of the contrast-enhanced imaging device shown in FIG. 1. It is a characteristic view which shows the luminance distribution of 1 line data by an imaging device. It is a correction curve figure which shows an example of contrast emphasis correction | amendment by the contrast emphasis imaging device shown in FIG. It is a correction curve figure which shows the other example of contrast emphasis correction | amendment by the contrast emphasis imaging device shown in FIG. It is a correction curve figure in case the gamma coefficient by the contrast emphasis imaging device shown in FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Imaging device 10 Imaging part 10a Lens 10b Image pick-up element 11 Image | video input part 12 Correction amount calculation part 12a 1 line data reading part 12b Maximum point detection part 12c Minimum point detection part 12d Intermediate | middle point detection part 12e Bandwidth calculation part 12f Coefficient calculation part 12g coefficient holding unit 13 video correction unit 14 video output unit

Claims (4)

A video input unit for storing video from the imaging unit, a correction amount calculation unit for calculating a contrast correction amount from video luminance information of the video input unit, and a video for video correction with the correction amount calculated by the correction amount calculation unit In the contrast-enhanced imaging device including a correction unit and a video output unit that outputs a video image corrected by the video correction unit, the correction amount calculation unit includes a data reading unit that reads a predetermined amount of data from the video input unit; The local maximum point detection unit and the local minimum point detection unit for detecting the local maximum luminance value and the local minimum luminance value from a predetermined amount of data, and the detected local maximum point luminance value and local minimum point luminance value are adjacent to each other. midpoint divides the value and the minimum point brightness values into a plurality of blocks as one block, calculates the midpoint luminance value from the maximum point luminance value and minimum point luminance values in respective divided blocks The output unit, the bandwidth calculation unit for calculating the bandwidth from the maximum point luminance value and the minimum point luminance value in each block, and the contrast enhancement coefficient for contrast enhancement from each of the intermediate point luminance values and each band width, respectively. A contrast-enhancing imaging apparatus, wherein the image correction unit performs image correction for each block based on a contrast enhancement coefficient of the coefficient calculation unit.   2. The data reading unit according to claim 1, wherein the data reading unit reads at least one line data from the video input unit, and the intermediate point calculation unit includes one maximum point luminance value and one line in the pixel direction of the one line data. A contrast-enhanced imaging apparatus, characterized in that an intermediate point luminance value is calculated from a maximum point luminance value and a minimum point luminance value in a block that is divided by having a minimum point luminance value.   2. The data reading unit according to claim 1, wherein the data reading unit sequentially reads from the video input unit for each line data, and the intermediate point calculation unit includes one maximum point luminance value in the pixel direction and one line data for each line data. A contrast-enhanced imaging apparatus, characterized in that an intermediate point luminance value is calculated from a maximum point luminance value and a minimum point luminance value in a block segmented with one minimum point luminance value.   2. The intermediate point calculation unit according to claim 1, wherein the intermediate point calculation unit does not output an intermediate point luminance value when a difference between adjacent maximum point luminance values and minimum point luminance values in the block is equal to or less than a predetermined first set value. A contrast-enhanced imaging device characterized by comprising:

Images