Color deviation correction method and image correction apparatus

Abstract

A color deviation correction method and an image correction device are disclosed. The color deviation correction method includes finding a color deviation area in a detected image, analyzing the detected image to estimate a correction color value of the color deviation area, and modifying the color deviation area with the correction color value to generate a corrected detected image without color deviation.

Description

[0001] This application claims priority to U.S. Patent Application No. 17 / 585,603, filed January 27, 2022, entitled “Color Deviation Correction Method and Image Correction Device Thereof,” the entire contents of which are incorporated herein by reference. Technical Field

[0002] This invention provides a color deviation correction method and an image correction device, particularly a color deviation correction method and an image correction device that can be used to eliminate color fringing. Background Technology

[0003] With technological advancements, high-resolution imaging has become widely used in various types of imaging devices due to its richer image detail. However, many undesirable artifacts are becoming increasingly apparent in high-resolution images. These artifacts, known as "color fringing," detract from the aesthetics of the image. Color fringing manifests as out-of-focus purple ghosting around high-frequency areas of the image or edges of captured objects. Its primary cause is color distortion resulting from chromatic aberration or halos caused by lenses within the imaging device. Color fringing is particularly noticeable at the boundaries of objects with significant differences in color brightness. Traditional purple fringing removal techniques require manual operation by the user, such as manually marking the color-fringed areas in the image and manually specifying that these areas should be replaced with gray. Therefore, designing a color deviation correction technology that can automatically detect and repair color-fringed areas has become a key development goal for the imaging equipment industry. Summary of the Invention

[0004] The present invention provides a color deviation correction method and image correction device for eliminating color fringing, thereby solving the above-mentioned problems.

[0005] This invention discloses a color deviation correction method applied to an image detector. The method includes locating a color deviation region within a detected image, analyzing the image to estimate a corrected color value for the color deviation region, and modifying the color deviation region with the corrected color value to generate a corrected detected image without color deviation.

[0006] Optionally, the color deviation correction method further includes analyzing the color distribution of the detection image to estimate the corrected color value. Furthermore, the color deviation correction method further includes using object recognition technology to locate a specific object within the detection image, determining whether the color deviation area matches the pattern range of the specific object, and if the color deviation area matches the pattern range, analyzing the color distribution of the specific object to estimate the corrected color value. If the color deviation area does not match the pattern range, analyzing the color distribution of the detection image to estimate the corrected color value.

[0007] Optionally, the color deviation correction method further includes removing pixel values ​​from the color deviation area to create a mask, and filling the mask with the corrected color value to generate the corrected detection image. Furthermore, the color deviation correction method further includes obtaining multiple coordinate points of the color deviation area, and replacing the initial pixel values ​​of the multiple coordinate points with the corrected color value to generate the corrected detection image.

[0008] The present invention also discloses an image correction device, including an image detector and a processing unit. The image detector is used to acquire a detection image. The processing unit is electrically connected to the image detector. The processing unit is used to locate a color deviation region within the detection image, analyze the detection image to estimate the corrected color value of the color deviation region, and use the corrected color value to modify the color deviation region to generate a corrected detection image without color deviation.

[0009] Optionally, the processor further analyzes the color distribution of the detected image to estimate the corrected color value. The processor further uses object recognition technology to locate a specific object within the detected image and determines whether the color deviation area matches the pattern range of the specific object. If the color deviation area matches the pattern range, the processor analyzes the color distribution of the specific object to estimate the corrected color value. If the color deviation area does not match the pattern range, the processor analyzes the color distribution of the detected image to estimate the corrected color value.

[0010] Optionally, the processor further removes pixel values ​​from the color deviation area to create a mask, and fills the mask with the corrected color value to generate the corrected detection image. The processor further obtains multiple coordinate points of the color deviation area and replaces the initial pixel values ​​of the multiple coordinate points with the corrected color value to generate the corrected detection image.

[0011] The color deviation correction method and image correction device of this invention, after identifying color deviation areas within a detected image, modify the color deviation areas to correct color values ​​to present the original expected color of the detected image. This invention can utilize machine learning algorithms or any other color analysis algorithm that can achieve the same effect to search for color deviation areas and estimate the corrected color values ​​for those areas. Therefore, the color deviation correction method of this invention can automatically find and correct color deviation areas in a detected image, restoring the true colors of objects within the detected image, thus enabling the image correction device and its associated image capturing device to achieve better image presentation results. Attached Figure Description

[0012] Figure 1 This is a functional block diagram of the image correction device according to an embodiment of the present invention.

[0013] Figure 2This is a flowchart of a color deviation correction method according to an embodiment of the present invention.

[0014] Figures 3 to 5 This is a schematic diagram of the detected images at different processing stages according to an embodiment of the present invention.

[0015] Figure label:

[0016] 10 - Image correction equipment; 12 - Image detector; 14 - Processor. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] like Figure 1 and Figure 2 , Figure 1 This is a functional block diagram of the image correction device 10 in an embodiment of the present invention. Figure 2 This is a flowchart of the color deviation correction method in an embodiment of the present invention. The image correction device 10 is used to determine whether the detection image acquired by the image capturing device has a color distortion phenomenon, and automatically repairs it when a color distortion phenomenon is detected. The image capturing device can be installed indoors or outdoors. The image correction device 10 can be independent of the image capturing device or a part of the image capturing device. The image correction device 10 may include an image detector 12 and a processing unit 14. The image detector 12 can capture the detection image itself or acquire the detection image from an external device via wired or wireless transmission. The processing unit 14 is electrically connected to the image detector 12 and is used to execute the color deviation correction method of the present invention. Figure 2 The color deviation correction method described above is applicable to Figure 1 The image correction device 10 shown.

[0019] like Figures 2 to 5 , Figure 3 , Figure 4 as well as Figure 5 This is a schematic diagram of the detection image I at different processing stages in an embodiment of the present invention. First, the color deviation correction method can execute step S100 to obtain the detection image I and find the color deviation region F within the detection image I. The color deviation region F is a color edge with a specific color that appears at the boundary between the high-brightness area and the low-brightness area during color light imaging. For example... Figure 3As shown, if a red chair is placed on an off-white floor, a purplish-red color deviation area F will appear at the junction of the chair back and the floor (represented by diagonal stripes). Generally, step S100 searches for the color deviation area F within the entire detection image I; however, practical applications are not limited to this, and its variations may depend on design requirements.

[0020] For example, the detection image I can be further delineated into a region of interest (not shown in the attached figure) manually or automatically. Step S100 only searches for the color deviation area F within the region of interest to reduce computation time and improve computational efficiency. Manual delineation means that the user can use an input interface such as a keyboard or mouse to delineate the region of interest within a specific area of ​​the detection image I. This specific area is usually an area that the user subjectively considers to be prone to color fringing, or a key observation area where color difference is unacceptable. Automatic delineation allows the image correction device 10 to determine which area of ​​the detection image I has more frequent object changes, or whether there is a specific object present in the detection image I, thereby determining the position of the region of interest within the detection image I.

[0021] Next, step S102 is executed to analyze the detection image I to estimate the corrected color value of the color deviation area F. In step S102, the color deviation correction method may analyze the color distribution of the entire detection image I, for example, by using edge detection to find lines and color changes that may appear with color edges within the detection image I, and then estimating the corrected color value of the color deviation area F based on the color change trend. Alternatively, the color deviation correction method may also use object recognition technology to first find a specific object within the detection image I, and then determine whether the position of the color deviation area F matches the pattern range of the specific object. For example, if the specific object is a red chair placed on a beige floor, and the position of the color deviation area F falls at the boundary between the red chair and the beige floor, then the color deviation correction method can determine that the color deviation area F matches the pattern range of the specific object, and therefore the color distribution of the specific object can be analyzed to obtain the corrected color value of the color deviation area F. If the color deviation area F is determined to be outside the pattern range of a specific object, for example, if the location of the color deviation area F is not at the boundary between the red chair and the off-white floor, the color deviation correction method can be changed to analyze the color distribution of the entire detection image I to obtain the corrected color value of the color deviation area F.

[0022] Next, step S104 can be optionally executed to remove the pixel values ​​of the color deviation area F to create a mask M. As shown in Figure 4, the mask M is drawn with a diamond pattern to represent unoccupied pixels. Finally, step S106 is executed to determine the position and extent of the mask M, and then the corrected color values ​​of the color deviation area F are filled into the mask M to generate the corrected detection image I'; as shown in Figure 4. Figure 5As shown, the corrected detection image I' has no purplish-red color deviation area F at the junction of the chair back and the floor (the absence of diagonal stripes indicates color fringing elimination). If the color deviation correction method chooses not to execute step S104, the color deviation correction method can record the coordinate positions of multiple coordinate points of the color deviation area F, and then in step S106, replace the initial pixel values ​​of multiple coordinate points (i.e., the incorrect color of the color fringing) with the corrected color values ​​estimated in step S102, thus generating the corrected detection image I'.

[0023] In summary, the color deviation correction method and image correction device of the present invention, after identifying color deviation areas within a detected image, modify the color deviation areas to correct color values ​​to present the original expected color of the detected image. The present invention can utilize machine learning algorithms or any other color analysis algorithm that can achieve the same effect to search for color deviation areas and estimate the corrected color values ​​for those areas. Therefore, the color deviation correction method of the present invention can automatically find and correct color deviation areas in a detected image, restoring the true colors of objects within the detected image, thus enabling the image correction device and its associated image capturing device to achieve better image presentation results.

[0024] Obviously, those skilled in the art can make various modifications and variations to the embodiments of the present invention without departing from the spirit and scope of the invention. Therefore, if these modifications and variations fall within the scope of the claims of the present invention and their equivalents, the present invention also intends to include these modifications and variations.

Claims

1. A color deviation correction method, applied to an image detector, characterized in that, The color deviation correction method includes: Find a color deviation area in a detection image, wherein the color deviation area is a color edge with a specific color that appears at the boundary between the high brightness area and the low brightness area when the color light is imaged; The detected image is analyzed to estimate a corrected color value for one of the color deviation regions, wherein the color distribution of the detected image is analyzed to estimate the corrected color value; And modify the color deviation area with the corrected color value to generate a corrected detection image without color deviation; Also includes: Using object recognition technology to locate a specific object within the detected image; Determine whether the color deviation area matches the pattern range of the specific object; And if the color deviation area conforms to the pattern range, the color distribution of the specific object is analyzed to estimate the corrected color value.

2. The color deviation correction method according to claim 1, characterized in that, If the color deviation area does not match the pattern range, the color distribution of the detected image is analyzed to estimate the corrected color value.

3. The color deviation correction method according to claim 1, characterized in that, Also includes: Remove pixel values ​​from the color deviation area to create a mask; And the corrected color values ​​are filled into the mask to generate the corrected detection image.

4. The color deviation correction method according to claim 1, characterized in that, Also includes: Obtain multiple coordinate points of the color deviation area; And the initial pixel values ​​of the plurality of coordinate points are replaced with the corrected color values ​​to generate the corrected detection image.

5. An image correction device, characterized in that, These include: An image detector is used to acquire a detection image; The system also includes a processing unit electrically connected to the image detector. The processing unit is used to locate a color deviation region within the detected image. The color deviation region is a color edge with a specific color that appears at the boundary between a high-brightness area and a low-brightness area during color light imaging. The processing unit analyzes the detected image to estimate the corrected color value of the color deviation region. The processing unit further analyzes the color distribution of the detected image to estimate the corrected color value, and uses the corrected color value to modify the color deviation region to generate a corrected detected image without color deviation. The processing unit further uses object recognition technology to find a specific object in the detection image, and determines whether the color deviation area matches the pattern range of the specific object. When the color deviation area matches the pattern range, it analyzes the color distribution of the specific object to estimate the corrected color value.

6. The image correction device according to claim 5, characterized in that, When the processing unit determines that the color deviation area does not conform to the pattern range, it analyzes the color distribution of the detected image to estimate the corrected color value.

7. The image correction device according to claim 5, characterized in that, The processing unit further removes pixel values ​​from the color deviation area to create a mask, and fills the mask with the corrected color value to generate the corrected detection image.

8. The image correction device according to claim 5, characterized in that, The processing unit further obtains multiple coordinate points of the color deviation area and replaces the initial pixel values ​​of the multiple coordinate points with the corrected color values ​​to generate the corrected detection image.

Images