TW202314643A - Method and system for automatically searching for multi-grayscale regions of interest - Google Patents

Abstract

The present invention is related to a method and a system for automatically searching for multi-grayscale regions of interest. In the method, a digital image is generated through an image device that captures a picture of a standard multi-grayscale chart. The brightness information of the digital image is used to determine a first brightness threshold and a second brightness threshold. The brightness thresholds are used to binarize the digital image and obtain an overall binarized image. A grayscale region can be obtained from the overall binarized image. The region of interest indicates multiple grayscale blocks that are obtained from the grayscale region of the digital image, and the multiple grayscale blocks can be used to evaluate the image device.

Description

Method and system for automatically searching multi-grayscale regions of interest

The present invention relates to a method for searching a region of interest in an image, in particular to a method and system for automatically searching for a multi-grayscale region of interest for evaluating grayscale information in an image.

The main components of a digital camera include optical systems (such as lenses and optical components) and circuit systems. The circuit systems mainly include image sensors, image processors and control circuits, and image sensors such as charge-coupled devices (CCDs). Complementary metal-oxide-semiconductor (CMOS), the control circuit is used to control the aperture, shutter, photosensitive parameters and camera program, and the image processor is used to process the image data formed from the image sensor , resulting in the final image.

Taking imaging equipment as an example, the process of forming a digital image is that the image sensor senses the light obtained through the lens according to various photosensitive parameters, and then generates image data, and the digital image is generated by the image processor. Generate digital images to evaluate imaging equipment.

Imaging equipment such as cameras and video cameras will evaluate the quality of the equipment before leaving the factory. It is more sensitive to grayscale differences, and the color shift of grayscale images in the color expression process is an important factor affecting image quality. Therefore, image correction needs to refer to image grayscale information, and the conventional detection method is to manually match detection equipment Determining image quality is prone to errors or inconsistent detection quality.

In view of the absence of the conventional methods for detecting grayscale information in images, the disclosure of the present invention discloses a method and system for automatically searching multi-grayscale ROIs.

In one embodiment, the method for automatically searching for a multi-gray-scale region of interest can be implemented in a computer system, and can be applied to evaluate imaging equipment. In the method, the imaging device is first obtained to shoot a digital image produced by a standard multi-grayscale image, for example, the standard multi-scale grayscale image is a Kodak multi-grayscale graphics card, and then the first brightness is obtained according to the brightness information in the digital image Preferably, the threshold and the second brightness threshold can be the brightest value and the darkest value obtained from the digital image.

The digital image is then binarized with the first brightness threshold to obtain a first binary image, and the digital image is binarized with a second brightness threshold to obtain a second binary image, and the first binary image is synthesized The first binarization map and the second binarization map can be used to obtain a total binarization map, and then a grayscale image area in the digital image can be determined from the total binarization map. This part can be determined from the total binarization map first. An area with a width and a length can define a grayscale image area.

Then, according to the area of the grayscale map, the region of interest including multiple grayscale blocks can be obtained from the digital image. The method can use a circle division method or a series algorithm to divide multiple grayscale areas from the digital image. block, the image device can be evaluated after comparing the gray scale value of each gray scale block in the multiple gray scale blocks with the standard multi-scale gray scale image.

In order to further understand the features and technical content of the present invention, please refer to the following detailed description and drawings related to the present invention. However, the provided drawings are only for reference and description, and are not intended to limit the present invention.

In order to fully understand the purpose, features and effects of the present invention, the present invention will be described in detail through the following specific embodiments and accompanying drawings, as follows:

In the present disclosure, "a" or "an" is used to describe the elements, elements and components described herein. This is done for convenience of description only and to provide a general sense of the scope of the invention. Accordingly, unless it is obvious that it is otherwise indicated, such description should be read to include one, at least one, and the singular also includes the plural.

As used herein, the terms "comprises", "including", "has", "containing" or any other similar terms are intended to cover a non-exclusive inclusion. For example, an element, structure, article, or device that contains a plurality of elements is not limited to those elements listed herein, but may include elements that are not explicitly listed but are generally inherent in the element, structure, article, or apparatus. other requirements. In addition, unless expressly stated to the contrary, the word "or" means an inclusive "or" and not an exclusive "or".

The present invention relates to a method for automatically searching a multi-gray-scale region of interest and a system for implementing the method. For the application situation, please refer to the schematic diagram of a system embodiment for implementing the method shown in FIG. 1 , in which the computer system 10 proposed by the display system is shown. , the computer system 10 is connected with an image device 100 to be evaluated, such as a camera and a video camera. In the method, the imaging device 100 shoots a standard multi-scale grayscale image, and uses the digital image output by the imaging device 100 to compare the captured standard multi-scale grayscale image to evaluate the imaging device 100, or as a calibration method of the imaging device 100. in accordance with.

The computer system 10 is provided with an image capture unit 101, such as an image capture card or an image capture device, which receives the digital image output by the imaging device 100 through the operation of its software and hardware, and then uses the processing unit 103 to process the received digital image , especially to obtain the region of interest in the digital image, and evaluate the imaging capability of the image capture device according to the result of searching the region of interest. In the above-mentioned embodiment, the multi-level gray scale information is acquired, compared with the image information of the standard multi-level gray scale image, and the evaluation result is output through the output unit 105 .

Furthermore, in the method proposed by the invention, the standard multi-level grayscale image captured by the imaging device 100 needs to have a gray area, such as a Kodak multi-grayscale graphics card (Kodak grayscale chart), or the background can be entirely gray, For white or black, you can obtain the specific brightness information of the screen, such as the average brightness, or the brightest and darkest values, and set it as the threshold value, and then use morphology to find the gray-scale area, and test it out through the image output of the imaging device. Noise generated by photosensitive elements in imaging equipment can be used to evaluate its image stability. The disclosed method for automatically searching for multi-gray-scale regions of interest mainly uses multi-gray-scale graphics in images as regions of interest, and one of its technical purposes is to check or correct the digital output of imaging equipment through the multi-gray-scale information in the regions of interest. The image can be used to evaluate the image device, and the image device can also be calibrated using the above-mentioned standard multi-level grayscale image.

FIG. 2 shows a flow chart of an embodiment of a method for automatically searching a multi-grayscale region of interest.

Prepare an image device, which can be any image system that can capture and output images, and prepare a standard multi-level gray scale image, as shown in the example in FIG. 3 . At the beginning of the method, a standard multi-scale grayscale image is captured by an imaging device (step S201) to generate a digital image, and after being obtained by a computer system, an area of interest is extracted from the digital image, such as a grayscale image area, After capturing the information, compare it with the standard multi-level gray scale image to evaluate the imaging quality of the imaging equipment.

Using the image processing method in the computer system to obtain the pixel value in the image, and obtain the brightness information of the image from the pixel value, the method is to determine the threshold of the subsequent calculation of binarization (binarization), according to one of the embodiments , determine the first brightness threshold and the second brightness threshold according to the brightness distribution in the digital image, for example, the average brightness of the image can be obtained from the pixel value (such as 0 to 255), so as to determine the brightness threshold, or get the brightness threshold from the image Find the brightest and darkest values (step S203), the first brightness threshold can be set to the brightest value in the digital image, and the second brightness threshold can be set to the darkest value in the digital image.

Afterwards, the method is used to binarize the digital image with the first brightness threshold to obtain the first binarized image (step S205). Refer to FIG. 4 for an embodiment diagram of obtaining the binarized image with the brightest threshold value. , the figure shows that the image pixels after binarization calculation according to the brightness threshold are left with white and black areas represented by 0 and 1 respectively; then the digital image is binarized by the second brightness threshold to obtain the second binarization For the figure (step S207 ), please refer to the figure of an embodiment of obtaining the binarized figure with the darkest threshold value shown in FIG. 5 , which also shows a figure with white and black areas after binarization calculation based on the brightness threshold. Next, two binarized images are synthesized (step S209 ). Refer to the embodiment diagram of the total binarized image formed by the brightest and darkest threshold values shown in FIG. 6 .

According to the total binarized image synthesized by the above steps, the grayscale image in the image can be judged by image processing technology. For example, in step S211, the length and width of the image can be determined according to the brightness threshold set in the above steps. . What is mentioned here is that in this step, Morphology can be used to find interesting areas in the image. The morphology is a nonlinear image analysis technology that can be used to analyze and identify specific areas and graphics in the image. For example, specific boundaries and contours can be obtained through morphology. In the method disclosed in the present invention, the main purpose is to obtain the grayscale image area formed by shooting the standard multi-scale grayscale image in step S201 from the digital image. . For example, morphological processing can be performed on the total binarized image obtained in step S209, and operations such as erosion and dilation are sequentially performed on the edges of the brightest and darkest areas in the image to eliminate noise , and after considering the collection of various scattered areas, the length and width of the grayscale image area can be accurately analyzed.

The image region obtained by the morphology can be compared with the region of interest to be obtained by the method, that is, the grayscale image region with a width and a length in the digital image determined by the total binarization image. Afterwards, a circle division method or a first-order algorithm can be used to divide a plurality of gray-scale blocks from the gray-scale map area in the digital image, which is the region of interest (ROI) to be obtained by the method. (step S213). The region of interest can be obtained from a digital image as shown in FIG. 7 with reference to an embodiment diagram comprising a plurality of gray-scale blocks. Through the above-mentioned circle division method or series algorithm, it can be further determined by the total binarization diagram. A grayscale area with a width and a length is divided into multiple grayscale blocks, and adjacent grayscale blocks have a certain difference in grayscale value, so the grayscale of each block can also be calculated by image processing methods One of the purposes is to compare the grayscale value of each grayscale block with the standard multi-scale grayscale image taken initially (such as step S201), so as to evaluate the imaging equipment that produces digital images (step S217 ).

Figure 8 shows a schematic diagram of dividing multiple grayscale blocks from a digital image by using the circle division method as an example. The purpose of using the circle division method is that when the image equipment takes pictures, the lens will have distortion (distortion), through the figure 8 The circle division method schematically indicated can correctly divide multiple gray-scale blocks according to the gray-scale information of pixels by considering the curvature of the lens, that is, there are system-set differences between adjacent gray-scale blocks, so that the circle can be The division method finds the interval between the gray scales according to the law of proportion, and obtains multiple gray scale blocks as shown in the figure.

What's more, the method can also divide a plurality of gray-scale blocks from the gray-scale map area in the digital image by a series algorithm, such as a geometric series, an arithmetic series or a Taylor series, etc., related The series can refer to the following function, which also considers the distortion caused by the curvature of the lens of the imaging device, which can be corrected exponentially from the center of the lens outward, thereby dividing each gray-scale block.

n = order (step)

r = deformation value (distortion)

a = width of grayscale

To sum up, according to the method and system for automatically searching for multi-gray-scale regions of interest described in the above embodiments, compared with the current method of manually marking regions of interest, the proposed method proposes a smart automatic search for regions of interest The method, combined with the binarization algorithm, morphology, and the algorithm for dividing grayscale blocks, automatically searches for the region of interest in the image from the image obtained by shooting the standard multi-scale grayscale image taken by the imaging equipment. One of its purposes can be through Grayscale information in images evaluates imaging devices.

The present invention has been disclosed above with preferred embodiments, but those skilled in the art should understand that the embodiments are only used to describe the present invention, and should not be construed as limiting the scope of the present invention. It should be noted that all changes and substitutions equivalent to the embodiment should be included in the scope of the present invention. Therefore, the scope of protection of the present invention should be defined by the scope of the patent application.

100: Imaging equipment 10:Computer system 101: Image capture unit 103: Processing unit 105: output unit S201～S217: Steps

Fig. 1 shows the schematic diagram of the system embodiment of the method for automatically searching for multi-gray-scale regions of interest; Fig. 2 shows the flow chart of an embodiment of the method for automatically searching for a multi-grayscale region of interest; Figure 3 shows a standard multi-level grayscale image; Fig. 4 shows the embodiment diagram of obtaining the binarization map with the brightest threshold value; Fig. 5 shows the embodiment diagram of obtaining the binarization map with the darkest threshold value; Fig. 6 shows the embodiment diagram of the total binarization map formed by the brightest and darkest threshold values; FIG. 7 shows a diagram of an embodiment of obtaining a region of interest including a plurality of grayscale blocks from a digital image; and FIG. 8 is a schematic diagram of an embodiment of dividing a plurality of gray scale blocks from a digital image by using a circle division method.

S201-S217: Method

Claims (10)

A method for automatically searching for a multi-grayscale region of interest, comprising: Obtaining a digital image produced by shooting a standard multi-level gray scale image; obtaining a first brightness threshold and a second brightness threshold according to the brightness information in the digital image; binarizing the digital image with the first brightness threshold to obtain a first binarized map, and binarizing the digital image with the second brightness threshold to obtain a second binarized map; synthesizing the first binarized map and the second binarized map to obtain a total binarized map; determining a gray scale image area in the digital image from the total binarized image; and According to the area of the gray scale image, the region of interest including a plurality of gray scale blocks is obtained from the digital image. The method for automatically searching for a multi-gray-scale region of interest as described in Claim 1, wherein the first brightness threshold is a brightest value in the digital image; the second brightness threshold is a darkest value in the digital image . The method for automatically searching for a multi-grayscale ROI as described in Claim 1, wherein the standard multi-grayscale image is a Kodak multi-grayscale graphics card. The method for automatically searching multi-gray-scale regions of interest according to claim 1, wherein the gray-scale image region with a width and a length is determined from the total binarized image. The method for automatically searching for a multi-gray-scale region of interest as described in any one of claim items 1 to 4, wherein the gray-scale map area obtained from the digital image is obtained by using a circle division method or a first-order arithmetic algorithm Divide the plurality of gray scale blocks. The method for automatically searching for a multi-grayscale region of interest as described in claim item 5, wherein the grayscale value of each grayscale block in the plurality of grayscale blocks is compared to the standard multi-scale grayscale image to evaluate An imaging device that generates the digital image. A system for automatically searching for a multi-grayscale region of interest, including: A computer system includes an image capture unit, a processing unit, and an output unit, wherein a method for automatically searching for a multi-gray-scale region of interest is run, the method comprising: receiving a digital image generated by shooting a standard multi-scale grayscale image by an image device through the image capture unit; The processing unit processes the digital image to obtain a first brightness threshold and a second brightness threshold according to brightness information in the digital image; The processing unit binarizes the digital image with the first brightness threshold to obtain a first binarized map, and binarizes the digital image with the second brightness threshold to obtain a second binarized map; synthesizing the first binarization map and the second binarization map by the processing unit to obtain a total binarization map; and A grayscale image area in the digital image is determined by the processing unit from the total binarized image, and an interest area including a plurality of grayscale blocks is obtained from the digital image according to the grayscale image area. The system for automatically searching for a multi-grayscale region of interest as described in Claim 7, wherein the first brightness threshold is a brightest value in the digital image; the second brightness threshold is a darkest value in the digital image . The system for automatically searching multi-gray-scale regions of interest as described in claim 7, wherein the gray-scale image region with a width and a length is determined from the total binarized image. The system for automatically searching for a multi-gray-scale region of interest as described in any one of claim items 7 to 9, wherein the gray-scale map area obtained from the digital image is obtained by using a circle division method or a first-order arithmetic algorithm Divide the plurality of gray scale blocks, and then compare the standard multi-scale gray scale image according to the gray scale value of each gray scale block in the multiple gray scale blocks, so as to evaluate an imaging device that generates the digital image , an evaluation is output through the output unit.

Images