JPH09284542A - Image input device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain an excellent synthesis image independently of property of an input image by selecting an area where saturated picture elements in image data of overlapped parts are small and accumulated luminance is high so as to calculate a luminance correction value. SOLUTION: An input image is processed in various signal processing ways and the processed image is written in image storage circuits 13, 14. Then a saturated picture element discrimination circuit 33 of an area selection circuit 16 discriminates saturated picture elements in each luminance correction detection area, accumulated sum circuits 31, 32 calculate the accumulate luminance values of the picture elements except the saturated picture elements and provide an output of the result to a discrimination circuit 34. Furthermore, a luminance correction detection area with highest one or both accumulated luminance value for each luminance correction detection area is selected and the accumulated luminance is outputted to an arithmetic circuit 21, in which a correction coefficient is calculated. Then the input image data are read from the image storage circuits 13, 14, gain adjustment circuits 22, 23 are used to correct the data and a synthesized image is outputted to an output terminal 19 via an output selection circuit 18.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention inputs an optical image of an object as image data corresponding to a plurality of images by a single or a plurality of image input sensors, synthesizes each image data to form one image. The present invention relates to an image input device to be formed.

[0002]

2. Description of the Related Art Conventionally, in an image input device, for the purpose of improving resolution, an optical image of a subject is divided and input from a single or a plurality of image input sensors, and the divided image data are combined to form one image. What constitutes an image is known. For example, when an optical image is captured using two image input sensors, the image data from the two image input sensors is used as 1
Although the image is constructed, as shown in FIG. 9, the first image input sensor captures the upper half image 101 and the second image input sensor captures the lower half image 102. 103 is the image 101
And the overlapping portion (overlapping area) of the image 102 are shown.

However, in the image input device having such a configuration, the signal level (hereinafter, brightness level) of each image data for the same image in the overlapping area is different due to the nonuniformity of the brightness characteristics of the two image input sensors. Happens. Therefore, even if the image data from the two image input sensors are switched at an appropriate portion of the overlapping area and combined into one image, the brightness level of the combined image data becomes a discontinuous step at the switching point. . This discontinuous step is CR
When output to a T display, a printer, etc., it appears as a line and the image quality is impaired.

As a means for solving such a problem, for example, as disclosed in Japanese Patent Laid-Open No. 63-299457, pixel data of an overlapping portion is read from an image of each area and a brightness correction amount is obtained. Therefore, it is known to correct the brightness. In (A) and (B) of FIG. 10, the areas of shaded regions 201 and 202 indicate the cumulative luminance value of the overlapping portion, and the correction coefficient is calculated by calculating the ratio of the cumulative luminance values of the two. I try to ask.

[0005]

However, according to the method disclosed in the above publication, as shown in FIGS. 11A and 11B, the image data from one image input sensor is saturated in the overlapping portion. Even if the region 303 is included, the ratio of the cumulative luminance value is calculated with the image data of only the shaded regions 301 and 302, which causes an error in the correction coefficient.
With respect to this point, nothing is taken into consideration in the technique disclosed in the above publication. In this case, there is a problem in that the error increases as the amount of data in the saturated region 303 that should be originally included in the calculation of the cumulative luminance value increases.

The present invention has been made in order to solve the above problems in the conventional image input apparatus. In the overlapping area, a saturated area in which data of a high brightness portion exceeding the sensitivity of the image input sensor is lost is formed. It is an object of the present invention to provide an image input device that can obtain a high quality composite image without deterioration of image quality even if it is included.

[0007]

In order to solve the above-mentioned problems, the invention according to claim 1 is characterized by a plurality of images in which spatially adjacent regions read by a single or a plurality of image input sensors overlap. In an image input apparatus for synthesizing image data corresponding to the above to form one image, area setting means for setting a plurality of areas in image data corresponding to an overlapping portion of the plurality of images, Of the image data of each area, an area selecting unit that selects one or a plurality of areas having a small number of saturated pixels in the saturation region of the sensitivity of the image input sensor and a large cumulative luminance value, and the area selecting unit. A brightness correction coefficient calculation means for detecting a brightness difference from the image data of the area and calculating a brightness correction coefficient is provided and configured.

In the image input device having such a configuration, a plurality of areas are set in the image data of the overlapping portion, and among the set image data of each area, the number of saturated pixels is small and the cumulative luminance value is large. Since the area is selected and the brightness correction value is calculated, the brightness correction coefficient with a small calculation error is calculated, and a good composite image with a small brightness difference in the overlapping portion is generated regardless of the characteristics of the input image. Obtainable.

According to a second aspect of the present invention, in the image input apparatus according to the first aspect, the area selection means includes saturated pixel determination means for determining saturated pixels in pixels in each area and removing the saturated pixels. A cumulative addition means for calculating a cumulative luminance value of pixels in each area excluding the saturated pixels is provided. By configuring the area selecting means in this way, the cumulative luminance value of each area excluding the saturated pixel whose luminance value is in the saturated state among the pixels in each area is calculated, and the area having the large cumulative luminance value is selected. By doing so, effective luminance ratio data with a large S / N can be obtained, and it is possible to calculate a luminance correction coefficient with a small error.

According to a third aspect of the present invention, in the image input apparatus according to the first aspect, the area selecting means sets the accumulated brightness value to a threshold value set in advance based on accumulated brightness value information for each area. It is configured to select the largest area among them. With this configuration, it is possible to select an area with a large cumulative luminance value while avoiding a saturated area in the overlapping portion of the images, it is possible to reduce the calculation error of the luminance correction coefficient, and to improve the characteristics of the input image. Regardless, a good composite image can be obtained.

[0011]

Next, an embodiment will be described. FIG. 1 is a block configuration diagram showing a first embodiment of an image input device according to the present invention. In this embodiment,
The invention corresponds to the invention described in claims 1 and 2. FIG.
2, 1 and 2 are a plurality of first image input sensors and a second image input sensor which are arranged so that spatially adjacent reading areas are overlapped, and 3 and 4 are from the image input sensors 1 and 2. An A / D converter for converting the image data into a digital signal, a signal processing unit 5 for performing a process of combining the image data from the A / D converters 3, 4 with each other, 6
Is an output signal terminal.

FIG. 2 is a diagram showing a configuration example of the signal processing unit 5. Reference numerals 11 and 12 are input signal terminals from the image input sensors 1 and 2, 13 and 14 are image storage circuits, 15 is a control circuit, and 16 is a control circuit. Is an area selection circuit, 17 is a brightness correction circuit, 18 is an output selection circuit for outputting two image data as one image signal, 19
Is an output signal terminal. Then, the brightness correction circuit 17
Is an arithmetic circuit 21 that calculates the ratio of the cumulative luminance values of the areas output from the area selection circuit 16 and gain adjustment circuits 22 and 23 that adjust the gain of the image data read from the image storage circuits 13 and 14. It is configured.

FIG. 3 is a diagram showing a configuration example of the area selection circuit 16 which constitutes the signal processing unit shown in FIG. 2. Reference numerals 31 and 32 are cumulative addition circuits, and 33 is a saturated pixel determination circuit for each pixel. The saturation value is determined by comparing the brightness value with a predetermined saturation level. Further, 34 is a determination circuit, 35 and 36 are image data input terminals, and 37 and 38 are cumulative luminance value output terminals.

Next, the operation of the first embodiment having such a configuration will be described. The input images captured by the first and second image input sensors 1 and 2 are A / D converted by the A / D converters 3 and 4, and the image storage circuits 13 and 14 of the signal processing unit 5 are processed.
Is written to. First and second image input sensors 1 and 2
As shown in FIG. 4, the adjacent input images 41 and 42 overlap with each other, that is, the adjacent images show the same image, and a portion 43 shown by a halftone dot is an overlapping portion of the images. .
Then, the overlapping portion of the image data of each of the input images 41 and 42
In 43, as shown in FIG. 5, a plurality of areas for detecting the brightness difference (hereinafter referred to as brightness correction detection areas) are set,
The control circuit 15 sequentially outputs the image data in each area to the area selection circuit 16. FIG. 5 is an example in which five brightness correction detection areas are set in the overlapping portion 43, and 51, 52, 53, 54, and 55 are the brightness correction detection areas belonging to one of the input images 41, and are shown in parentheses. Shown 61,6
Reference numerals 2, 63, 64, and 65 indicate respective brightness correction detection areas belonging to the other input image 42.

The area selection circuit 16 includes a saturated pixel determination circuit.
At 33, the saturated pixel of each luminance correction detection area is determined, and as shown in FIGS. 6A and 6B, the luminance correction detection area belonging to one input image and the other overlapping the input image. When the pixel is saturated in either one of the brightness correction detection areas belonging to the input image (in this example, the brightness correction detection area 51 belonging to one input image 41
7A and 7B include saturated pixels, and the brightness correction detection area 61 belonging to the other input image 42 does not include saturated pixels), and as shown in FIGS. It is assumed that the luminance level of both the pixel and the pixel of the other input image corresponding to the saturated pixel is 0 and does not exist. Then, in the cumulative addition circuits 31 and 32, the cumulative luminance values of the pixels excluding the saturated pixels are calculated and output to the determination circuit 34. This operation is performed for each set brightness correction detection area, and the calculation result is output to the determination circuit 34. In the determination circuit 34, one or both of the brightness correction detection areas have the largest cumulative brightness value, that is, the brightness correction detection area having the largest shaded area in FIGS. 7A and 7B. , And outputs the accumulated brightness value to the brightness correction circuit 17.

In the brightness correction circuit 17, the arithmetic circuit 21 calculates the ratio of the cumulative brightness value of the brightness correction detection area output from the area selection circuit 16, and calculates the correction coefficient. After that, the input image data is read from the image storage circuits 13 and 14, corrected by the gain adjustment circuits 22 and 23, and then the composite image is output to the output terminal 19 through the output selection circuit 18.

As described above, according to the present embodiment, when the luminance correction coefficient is obtained, the area selection circuit including the determination circuit of the saturated pixel is provided, so that the overlapped portion of the input image data is saturated. Even if it contains pixels
Even if it is not included, the brightness correction coefficient for matching the brightness characteristics of each input image data can be calculated accurately, so that a good composite image can be obtained.

Next, a second embodiment will be described. This embodiment corresponds to claims 1 and 3, FIG. 8 is a diagram showing a configuration of an area selection circuit in the second embodiment, and the configuration of other portions is the same as that of the first embodiment. Is the same as. The area selection circuit according to the second embodiment includes cumulative addition circuits 71, 72 and a threshold level setting circuit 7.
3 and a determination circuit 74, 75 and 76 are image data input terminals, and 77 and 78 are cumulative luminance value output terminals.

Next, the operation of the second embodiment will be described. First, as shown in FIG. 5, in the overlapping portion of each input image data, each brightness correction detection area 51, 52, 53, 54, 55 belonging to one input image for obtaining the brightness correction coefficient is obtained.
And the operation of setting each brightness correction detection area 61, 62, 63, 64, 65 belonging to the other input image overlapping with this and outputting the image data in the brightness correction detection area to the area selection circuit. The same as in the first embodiment.

In the area selection circuit, the cumulative addition value for each luminance correction detection area is calculated by the cumulative addition circuits 71 and 72. Subsequently, in the determination circuit 74, the threshold value set in the threshold level setting circuit 73 is compared with the cumulative brightness value, and the brightness correction detection area with the largest cumulative brightness value is selected among the threshold values in which the cumulative brightness value is set. The selected brightness is output to the brightness correction circuit 17 of FIG. 2 for the brightness correction detection area. In the calculation of the brightness correction coefficient, it is desirable that the larger the cumulative brightness value used for the calculation is, the larger the S / N is, but the larger the cumulative brightness value, the larger the probability of including the saturated pixel. For this reason, the cumulative brightness value at which the probability of including saturated pixels is reduced to some extent is set as the threshold value. Then, based on the accumulated brightness value of the selected brightness correction detection area, correction is performed as in the first embodiment.

Therefore, according to the present embodiment, in obtaining the brightness correction coefficient, by providing the area selection circuit to which the determination circuit of the cumulative brightness value is added, the probability that the saturated pixel is included in the overlapping portion of the input image is increased. Since a small brightness correction detection area having a large cumulative brightness value is selected and the ratio of the cumulative brightness values is calculated, a good composite image can be obtained.

In the first and second embodiments, the image input device is composed of a combination of circuits having respective operation functions, but of course it is composed of software having similar functions. It is also possible to do so. Further, in each of the above-described embodiments, a single area having the largest cumulative brightness value is selected as the brightness correction detection area, but a plurality of brightness correction detection areas may be selected. In this case, the amount of information increases, and the accuracy can be improved.

[0023]

As described above based on the embodiments, according to the invention described in claim 1, the luminance correction coefficient with a small calculation error is calculated, and the luminance of the overlapping portion is irrespective of the characteristics of the input image. A good composite image with a small difference can be obtained. According to the second aspect of the invention, since effective luminance ratio data with a large S / N can be obtained, it is possible to calculate a luminance correction coefficient with a small error. According to the third aspect of the invention, the calculation error of the brightness correction coefficient can be reduced, and a good composite image can be obtained regardless of the property of the input image.

[Brief description of drawings]

FIG. 1 is a block configuration diagram showing a first embodiment of an image input apparatus according to the present invention.

FIG. 2 is a diagram showing a configuration example of a signal processing unit in the embodiment shown in FIG.

3 is a diagram illustrating a configuration example of an area selection circuit in the signal processing unit illustrated in FIG.

FIG. 4 is a diagram showing an aspect of input images from two image input sensors in the embodiment shown in FIG.

5 is a diagram showing a brightness correction detection area set in an overlapping portion of the input image shown in FIG.

FIG. 6 is a diagram showing a brightness level of image data in a brightness correction detection area.

FIG. 7 is a diagram showing a mode in which the luminance levels of both the saturated pixel and the pixel corresponding to the saturated pixel are set to 0 in FIG. 6;

FIG. 8 is a diagram showing a configuration of an area selection circuit according to a second embodiment of the present invention.

FIG. 9 is a diagram showing an input mode of two input images.

FIG. 10 is a diagram showing luminance levels of image data in an overlapping portion of two input images.

[Fig. 11] Fig. 11 is a diagram illustrating a luminance level in a case where a saturated region exists in image data of an overlapping portion of two input images.

[Explanation of symbols]

 1, 2 Image input sensor 3, 4 A / D converter 5 Signal processing unit 6 Output signal terminal 11, 12 Input signal terminal 13, 14 Image storage circuit 15 Control circuit 16 Area selection circuit 17 Brightness correction circuit 18 Output selection circuit 19 Output signal terminal 21 Arithmetic circuit 22, 23 Gain adjustment circuit 31, 32 Cumulative addition circuit 33 Saturated pixel judgment circuit 34 Judgment circuit 35, 36 Input terminal 37, 38 Output terminal 41, 42 Input image 43 Overlapping part 51 ~ 55, 61 ~ 65 Brightness correction detection area 71, 72 Cumulative addition circuit 73 Threshold level setting circuit 74 Judgment circuit 75, 76 Input terminal 77, 78 Output terminal

Claims (3)

[Claims] 1. An image input device for forming one image by synthesizing image data corresponding to a plurality of images in which spatially adjacent regions are read by a single image input sensor or a plurality of image input sensors. An area setting means for setting a plurality of areas in image data corresponding to an overlapping portion in the plurality of images, and a saturated pixel in a saturated area of the sensitivity of the image input sensor in the image data of each set area Area selection means for selecting one or a plurality of areas having a small cumulative value and a large cumulative brightness value, and a brightness correction for detecting a brightness difference from image data of the area selected by the area selection means and calculating a brightness correction coefficient. An image input device comprising: a coefficient calculating unit. 2. The area selection means determines saturated pixels in pixels in each area to determine saturated pixel excluding the saturated pixels, and cumulative luminance values of pixels in each area excluding the saturated pixels. The image input device according to claim 1, further comprising cumulative addition means for calculating. 3. The area selecting means is configured to select an area having the largest cumulative luminance value below a preset threshold value based on the cumulative luminance value information for each area. The image input device according to claim 1, wherein the image input device is provided.