Color recognition method and device, equipment and storage medium

A color recognition and color classification technology, applied in the field of image processing, can solve the problem of color blindness or color weakness unable to distinguish natural spectral colors, etc., to achieve the effect of improving accuracy

Pending Publication Date: 2022-01-21

BAIDU ONLINE NETWORK TECH (BEIJIBG) CO LTD

0 Cites 0 Cited by

AI-Extracted Technical Summary

Problems solved by technology

Since color blindness or color weakness cannot distinguish at least some...

Method used

Embodiments of the present disclosure display the image to be recognized; in response to the region selection operation of the image to be recognized, determine the candidate display region according to the selected region; enlarge and display the corresponding region image of the candidate region to obtain a partially enlarged image; respond to the local A pixel selection operation for enlarging an image, identifying the target color of the selected pixel, and outputting the target color. By adopting the above technical solution, it is possible to provide color recognition assistance for people who cannot distinguish at least some colors, such as color blindness or color weakness, so that such people can perform color recognition without using a color blindness correction device, which reduces hardware costs and can Adapt to different types of user groups, with a wide range of adaptation. At the same time, the disclosure introduces the candidate area and enlarges the image of the corresponding area of the candidate area to obtain a partially enlarged image for pixel point selection, which facilitates the operator to accurately select pixel points, and avoids the recognition of errors caused by deviations in pixel point selection. The target color does not match the expected actual color, which improves the accuracy of color recognition results.

Embodiments of the present disclosure show the image to be recognized by the image display module to be recognized; the candidate region determination module responds to the region selection operation of the image to be recognized, and determines the candidate display region according to the selected region; enlarges the display candidate by enlarging the display module The region corresponds to the region image to obtain a partially enlarged image; the color recognition and display module responds to the pixel selection operation on the partially enlarged image to identify the target color of the selected pixel and output the target color. By adopting the above technical solution, it is possible to provide color recognition assistance for people who cannot distinguish at least some colors, such as color blindness or color weakness, so that such people can perform color recognition without using a color blindness correction device, which reduces hardware costs and can Adapt to different types of user groups, with a wide range of adaptation. At the same time, the disclosure introduces the candidate area and enlarges the image of the corresponding area of the candidate area to obtain a partially enlarged image for pixel point selection, which facilitates the operator to accurately select pixel points, and avoids the recognition of errors caused by deviations in pixel point selection. The target color does not match the expected actual color, which improves the accuracy of color recognition results.

It can be understood that by introducing preset color categories to divide the color categories in the selected area, the color situation contained in the selected area can be described qualitatively, providing data support for the determination of candidate areas, enriching How to determine the color category.

It can be understood that, by introducing the area area of the defined region of the pixel boundary of different color categories, the clipping reference area avoids the situation that the reference area is too large and the enlarged display effect is not good, and the reasonableness of the selected candidate area is improved. Therefore, it improves the rationality of the corresponding local enlarged image, facilitates the selection of pixels from the partially enlarged image, reduces the occurrence of misidentification of the color corresponding to the pixel, and thus...

Abstract

The invention provides a color recognition method and device, equipment and a storage medium, and relates to the technical field of image processing, in particular to the artificial intelligence technology. According to the specific implementation scheme, the method comprises the following steps: displaying a to-be-recognized image; in response to a region selection operation on the to-be-recognized image, determining a candidate region according to a selected region; magnifying and displaying a region image corresponding to the candidate region to obtain a locally magnified image; and in response to a pixel point selection operation on the locally amplified image, identifying a target color of the selected pixel point, and outputting the target color. According to the technology disclosed by the invention, the accuracy of a color recognition result is improved.

Application Domain

Geometric image transformationCharacter and pattern recognition

Technology Topic

Computer graphics (images)Imaging processing +2

Image

Examples

Experimental program(1)

Example Embodiment

[0026] The exemplary embodiments of the present disclosure will be described below, including various details of the embodiments of the present disclosure to facilitate understanding, and they should be considered simply exemplary. Accordingly, it will be appreciated by those skilled in the art that various changes and modifications can be made without departing from the scope and spirit of the disclosure. Also, for the sake of clarity and concise, the following description is omitted in the following description.

[0027] The various color identification methods and color identification devices provided herein are suitable for use in the application scenarios in color identification in the image in the image. The color identification methods provided by the present disclosure may be performed by a color identification device, which can be implemented using software and / or hardware and is specifically disposed in an electronic device. The electronic device can be in the terminal device or server, the present disclosure does not limit this.

[0028] For ease of understanding, the color identification method is first described in detail.

[0029] See figure 1 A color identification method shown, including:

[0030] S101, show the image to be identified.

[0031] Optionally, the image to be identified may be stored in advance, or in other storage devices associated with the electronic device; corresponding, acquisition and display of the image to be identified from the electronic device local or other storage device. Typically, it can be displayed from the electronic device local album to be displayed.

[0032] Alternatively, it is also possible to collect and display the image to be identified and displayed when there is a color identification requirement.

[0033] S102, in response to the area selection operation of the image to be identified, the candidate area is determined according to the selected area.

[0034] Among them, the area selection operation of the identified image can be a click operation, a box selection or drag operation, or the like, the disclosure is not limited. Candidate area is used to characterize the operational planned area. It should be noted that the present disclosure does not limit any particular shape of the candidate zone. Of course, in order to facilitate viewing, the shape of the candidate area can be set to a rule, such as a rectangular, circular or positive polygon.

[0035] Exemplary, according to the selected area, it is determined that the candidate area may be directly used as a candidate area; or, the selected area can be extended, and the extended area is used as a candidate area. Among them, the expansion parameters in accordance with the expansion may include at least one of the extension multiple and the expansion direction. The specific values of each extended parameter can be set by the skilled person based on the empirical value, or the operator is selected or set according to the actual needs.

[0036] It should be noted that when the regional selection operation cannot accurately characterize the operation of the operation, for example, when the image to be identified is displayed in the intelligent terminal, if the operator is clicked by a thicker touch, the image is clicked to be identified, resulting in a regional selection. When operating, it is possible to make a click area contains a certain deviation in the actual intention area, or, there is a certain deviation in the actual intent area of the area and the operator. Therefore, by the determination of the candidate zone, it is convenient for subsequent generation of local amplified images to facilitate accurate selection of pixel points to provide data support.

[0037] S103, an enlarged display of the candidate zone corresponding zone image to obtain a partially enlarged image.

[0038] The enlarged display candidate zone correspondence area image can be enlarged to enlarge the image to be identified, and the enlarged candidate zone is achieved; or only the candidate area correspondence area image, the present disclosure may not be limited to the specific enlarged object body.

[0039] In an alternative embodiment, the candidate zone is enlarged in accordance with the preset amplification ratio, to obtain a partially enlarged image, and show the partially enlarged image. Among them, the preset amplification ratio can be determined by the technician as needed or experienced, or by the operator.

[0040] Since the local amplification image is displayed based on the preset amplification ratio, there may be an enlarged result is not ideal (for example, too much or too small), so that the operating party cannot perform the selection or viewing discomfort of the pixel point of the desired identification; or require an operator Manually adjust the preset amplification ratio, reducing color identification efficiency. In order to avoid the above case, in another alternative embodiment, it is also possible to automate the determination of the target amplification ratio, and in accordance with the target amplification ratio, the candidate area corresponds to the area image to be enlarged.

[0041] Exemplary, the target amplification ratio can be determined according to the candidate amplification ratio according to the candidate zone; the candidate zone is enlarged according to the target amplification ratio, and is to obtain a partially enlarged image.

[0042] The various category colors are pre-set corresponding candidate amplifier ratios, and the values of each candidate amplifier ratio can be set by the skilled person, or by the operator according to their own requirements. It should be noted that the candidate amplification ratio corresponding to different categories color can be the same or different, and this disclosure does not limit this.

[0043] In a concrete implementation, the maximum amplification ratio, minimum amplification ratio, mean amplification ratio, median amplification ratio or random selection ratio, median amplification ratio or random selection ratio, median amplification ratio or random selection ratio Zoom in ratio.

[0044] Conveniently limited to electronic device itself and operates to operate the electronic device to operate, the local amplification display area is typically set in the electronic device, which is used to display partial enlarged images, thereby restricting partial enlarged image size, avoiding The local amplification image is too large, causing the operator to operate the electronic device unchanged while avoiding the local amplification image is too small, resulting in the following cases where the pixel point is selected. In order to determine the target amplification ratio while automation, the local enlarged image can adapt to the local amplification display area size, in another specific implementation, the reference zoom ratio can be determined according to the ratio of the candidate area and the local amplification display area; The reference magnification ratio is selected from the candidate amplifier ratio, and the target amplification ratio is selected from the candidate zone.

[0045] Specifically, the ratio of the candidate area and the local amplification display area can be used as a reference amplification ratio; from the candidate amplifier ratio, select the candidate magnification ratio of the reference amplification ratio difference (eg, minimum) in the candidate zone As the target amplification ratio. Of course, in order to avoid local amplifying images overflow local amplification display area, or partially enlarged display area cannot display the partial amplifying image, it is also possible to select and reference amplification ratios from each candidate amplification ratio than the reference amplification ratio. A candidate amplification ratio having a smaller value (for example, the smallest), as the target amplification ratio.

[0046]It will be appreciated that the above-mentioned optional embodiment will be partially enlarged to determine the determination process, refine to determine the target amplification ratio according to each type of color in the candidate area, and determine the target amplification ratio, and correspond to the candidate area corresponding to the area image according to the target zone ratio. Make an enlargement process to obtain a partially enlarged image. With the above technical solution, automation of the target amplification ratio can be achieved, and manual access is avoided, resulting in a lower color recognition efficiency. Meanwhile, the above technical solution can determine the target amplification ratio according to the candidate amplifier ratio, and the mechanical enlargement ratio is calculated in accordance with the candidate zone, and the colligation zone is enlarged in accordance with the fixed zoom ratio of the mechanical region. In case of excessive or too small, it improves the effectiveness of the resulting partially enlarged image.

[0047] In one alternative embodiment, after obtaining a local amplified image, the local enlarged image can be further updated in response to the zoom operation of the local amplified image.

[0048] The zoom operation can be implemented by touching the scaling function button, or performs preset zoom operations in a local enlarged image through the touch. For example, you can implement the distance between your fingers, to realize the distance between the double fingers, and realize the distance between the fingers, and realize the reduction function. Among them, the distance between the double fingers is associated with the magnification or reduction.

[0049] S104, in response to the pixel point selection operation of the local amplified image, identify the target color of the selected pixel point, and output the target color.

[0050] Among them, the pixel point selection operation of the local amplified image can be a click operation or a box selection operation, or the like, the present disclosure does not limit the specific operation mode of the pixel point selection operation.

[0051] Exemplary, identifying the target color of the selected pixel point, can determine the color value of the selected pixel point based on the color identification extractor; the color category belongs according to the color value, as the target color. Among them, the color value can be used with R (Red, red) G (Green, green) B (Blue, blue) color value, or hex color code, or may be H (Hue, color) s (Saturation, saturation) Degree) L (Lightness, brightness) color value is represented.

[0052] Exemplary, output target colors can be implemented in a text display and / or voice playback.

[0053] In an alternative embodiment, if the object color is output in a manner, the target color can be output, and any region other than the local amplification display area corresponding to the local amplification image can be displayed. To facilitate viewing, the color result display area is typically set in advance, and the color result is displayed in advance, and the text is displayed.

[0054] In another alternative embodiment, if the object color is output in the manner, the target color can be used, and the color of the target color different color system can be used in the local amplification display area of the local enlarged image. In order to facilitate find, you can also use the color of different colors of colors contained in the candidate area to perform text display. It should be noted that in order to avoid the color display to bring discomfort to the operator, you usually use the default color to make a text display, such as black or white, only different colors of the target color only when the default color is not easy to distinguish between the default colors. Other colors of the department perform text display.

[0055] The present disclosure is shown to be identified; in response to a region selected operation to be identified, the candidate display area is determined according to the selected area, and the aligns of the candidate area corresponds to the corresponding area image to obtain a partially enlarged image; in response to a local enlarged image Pixel point selection operation, identify the target color of the selected pixel point, and output the target color. With the above technical solution, it is possible to provide color identification help to a population that cannot distinguish at least some of the color, such as a color blind, or the weak, so that the class can make color identification without the need for a color blind correction device, reducing hardware cost, and capable Adapted to different categories of user bases, wide adaptation range. At the same time, the present disclosure is enlarged by introducing a candidate area, and is enlarged to the candidate zone corresponding to the region image, which is obtained to provide a local amplified image selected by the pixel point, which is convenient for the operator to accurately perform the pixel point selection, avoiding the deviation due to the pixel point selection. The target color, the actual color does not match the actual color of the expectation, improves the accuracy of color recognition results.

[0056] Based on the above technical solutions, the present disclosure also provides an alternative embodiment. In this alternative embodiment, an optimized improvement is performed on the determination process of the local amplified image. It should be noted that the representation of the foregoing embodiments can be referred to in the present disclosure embodiment.

[0057] See figure 2 A color identification method shown, including:

[0058] S201, display the image to be identified.

[0059] S202, in response to the area selection operation of the identified image, determines the candidate area according to the selected area.

[0060] S203, the identification image is enlarged to obtain a global amplified image.

[0061] S204, showing the corresponding area image of the candidate area in the global amplification image to obtain a partially enlarged image.

[0062] Exemplary, it can be enlarged according to the preset amplification ratio, and the identification image is enlarged, resulting in a global amplified image; only showcase the candidate area corresponding to the candidate area in the global amplification image, in order to facilitate explanation, the partial area image is referred to as partial enlargement image. Among them, the preset amplification ratio can be determined by the technician as needed or experienced, or by the operator.

[0063] Since the image to be enlarged according to the preset amplification ratio, the amplification result of the local amplified image displayed is not ideal (e.g., excessively or too small), so that the operation party cannot perform the selection or viewing of the desired identification pixel point or viewing discomfort. The situation occurs; or requires the operator to manually adjust the preset amplification ratio, reduce color identification efficiency. In order to avoid the above case, in another alternative embodiment, it is also possible to automate the determination of the target amplification ratio, and in accordance with the target amplification ratio, the candidate area corresponds to the area image to be enlarged.

[0064] Exemplary, the target amplification ratio is determined according to the candidate amplification ratio according to each category color in the candidate zone; the identification image is enlarged according to the target amplification ratio, resulting in a global amplified image; only the candidate area corresponding to the global amplified image Get partially enlarged images.

[0065] The various category colors are pre-set corresponding candidate amplifier ratios, and the values of each candidate amplifier ratio can be set by the skilled person, or by the operator according to their own requirements. It should be noted that the candidate amplification ratio corresponding to different categories color can be the same or different, and this disclosure does not limit this.

[0066] In a concrete implementation, the maximum amplification ratio, minimum amplification ratio, mean amplification ratio, median amplification ratio or random selection ratio, median amplification ratio or random selection ratio, median amplification ratio or random selection ratio Zoom in ratio.

[0067] Conveniently limited to electronic device itself and operates to operate the electronic device to operate, the local amplification display area is typically set in the electronic device, which is used to display partial enlarged images, thereby restricting partial enlarged image size, avoiding The local amplification image is too large, causing the operator to operate the electronic device unchanged while avoiding the local amplification image is too small, resulting in the following cases where the pixel point is selected. In order to determine the target amplification ratio while automation, the local enlarged image can adapt to the local amplification display area size, in another specific implementation, the reference zoom ratio can be determined according to the ratio of the candidate area and the local amplification display area; The reference magnification ratio is selected from the candidate amplifier ratio, and the target amplification ratio is selected from the candidate zone.

[0068] Specifically, the ratio of the candidate area and the local amplification display area can be used as a reference amplification ratio; from the candidate amplifier ratio, select the candidate magnification ratio of the reference amplification ratio difference (eg, minimum) in the candidate zone As the target amplification ratio. Of course, in order to avoid local amplifying images overflow local amplification display area, or partially enlarged display area cannot display the partial amplifying image, it is also possible to select and reference amplification ratios from each candidate amplification ratio than the reference amplification ratio. A candidate amplification ratio having a smaller value (for example, the smallest), as the target amplification ratio.

[0069] It will be appreciated that the above-mentioned optional embodiment will enlarge the determination process of the image globally, refine to determine the target amplification ratio according to each category color in the candidate area, and to enlarge the image according to the target amplification ratio. Get globally enlarged images. With the above technical solution, automation of the target amplification ratio can be achieved, and manual access is avoided, resulting in a lower color recognition efficiency. At the same time, the above technical solution can determine the target amplification ratio according to the candidate amplification ratio, and the mechanical amplification ratio is performed in accordance with the candidate amplification ratio, and the mechanical enlargement ratio is flexible. There is no excessive or too small, and the effectiveness of the resulting partial amplification image is improved.

[0070] In an alternative embodiment, the candidate area corresponding to the candidate zone in the global amplification image is shown only in the local enlarged display area, and the local amplification image is obtained, and the data support is provided for subsequent pixel points. In order to flexibly perform the selection of adjacent region pixel points, after obtaining a local amplification image, the following operations can also be added: in response to the mobile operation of the global amplified image, the local enlargement image is updated.

[0071] Among them, the mobile operation of the global amplified image can be implemented by pressing buttons control or drag, etc., the present disclosure does not limit this.

[0072] Specifically, when receiving the operation operation of the operator on the global amplifying image, in the local amplification image correspond to the local amplification display area, the local enlarged image is updated according to the moving direction and the moving length.

[0073] It will be appreciated that the pixel point selection of adjacent regions of the candidate area can be flexibly performing the pixel point selection of adjacent regions of the candidate zone, and improves the convenience of neighboring region pixel point color identification.

[0074] S205, in response to the pixel point selection operation of the local amplified image, identify the target color of the selected pixel point, and output the target color.

[0075] The present disclosure is refined to the determination operation of the local amplified image, and the global amplifying image is obtained; the global amplification image corresponds to the candidate area corresponding to the area image to obtain a partially enlarged image. The above technical solution is directly enlarged to the global amplification image, and insteads to enlarge the candidate zone corresponding to the region image, there is no need to perform the segmentation of the area image of the candidate area, so that the amplification process is more convenient, and there is no need to allocate memory space for the area image of the candidate area. Reduce the occupation of memory resources. Further, the integration of the identification image is enlarged, resulting in a global amplifying image, providing a selection of a pixel point of the adjacent area of the candidate area based on the global amplification image, providing a color identification in at least two regions in the same to be identified. In the case of, the color identification efficiency can be improved, and the calculation of the calculations determined by repeated candidate area determination and local amplification area can be reduced.

[0076] Based on the above technical solutions, the present disclosure also provides an alternative embodiment. In this alternative embodiment, an optimized improvement is performed on the determination process of the candidate area. It should be noted that the representation of the foregoing embodiments can be referred to in the present disclosure embodiment.

[0077] See image 3Color recognition method, including:

[0078] S301, display the image to be identified.

[0079] S302, in response to the regional selection operation of the image to be identified, determine the candidate area according to the color attribute of the selected area.

[0080] Where color properties are used to characterize the color distribution of the selected area. The color attribute can specifically include at least one of a color category, a color number, and a color value difference between adjacent colors.

[0081] In an alternative embodiment, according to the color attribute of the selected area, it is determined that the candidate area may be: If the preset color is included in the selected area, the selected area is expanded according to the preset color, and the selected area is expanded. Get the candidate area. The extended parameters may include at least one of the expansion or extension, and the like. Among them, preset colors and preset colors correspond to the extension parameters, can be pre-set by the technicians to choose from the operator.

[0082] In another alternative embodiment, the candidate area is determined according to the color attribute of the selected area, which may be: Identify the color category in the selected area; if the number of color categories in the selected area is larger than the preset number threshold, then choose The higher extension multiple is extended to the selected area; if the number of color categories in the selection area is not more than the preset number threshold, select a lower extension multiple to expand the selected area. Among them, the specific extension can be set by the technician as needed or experienced. The expansion direction can be set by the skilled person based on the empirical value, or by the operator according to the actual demand. For example, the expansion direction can be set to be centered in the selected area, and extension in the circumference direction.

[0083] It should be noted that the above-mentioned alternative embodiment is an example of a candidate area as an example. Of course, it is also possible to set at least two preset quantity thresholds according to the actual needs, and at least three extensions are set. Accordingly, the corresponding threshold interval is determined by the adjacent predetermined quantity threshold, and the corresponding relationship between different threshold intervals and extensions is established; wherein the number of threshold intervals having a higher value of the interval element corresponds to a higher value; interval A threshold interval having a lower element value corresponds to a lower value.

[0084] In yet another alternative embodiment, according to the color attribute of the selected area, the candidate area can be determined: identifying the color category in the selected area; determines the candidate area according to the color category, to enable the candidate zone to include each color category Corresponding pixel points. In a specific implementation, each color category in the candidate area is uniformly distributed.

[0085] In still another alternative embodiment, according to the color attribute of the selected area, the candidate area can be determined: identify the color category in the selected area, and determine the pixel boundary of each category color; according to the pixel boundaries of each category color. Determine the candidate area. The advantage of the above technical solution is that the candidate area is determined according to the distribution of the color category of the selected area and the pixel point extension, the candidate area is improved, and the admiring of the image to be identified to be identified is improved, which is in a local enlarged image. Accurate selection of pixel points provides protection.

[0086] Exemplary, identifying the color category in the selected area, which can be: Identify the preset color category included in the selected area, and determine the color category based on the identification result. Among them, the preset color category can be determined by the technician according to the empirical value, for example, a special population is easy to mix category, such as blue, green, red, etc .; the preset color category can also be divided according to different classification standards, such as by spectrum division For red, orange, yellow, green, blue, blue, and purple seven colors. Of course, the preset color category can also be divided in other ways, and this disclosure does not limit this.

[0087] It will be appreciated that by introducing the preset color category divides the color category in the selected area, it is possible to qualitatively describe the color conditions contained in the selected area, providing data support for the determination of the candidate area, enriched color category Determine.

[0088] Exemplary, identifying the color category in the selected area, which may be: the color value of the center pixel point of the selected area as the reference color value, and according to the color value of the non-center pixel point in the selected area and the reference color value Difference value, determine the color category.

[0089] Specifically, the geometry center of the selected area can be determined, and the pixel point located in the geometry center is used as the center pixel point; the color value of the center pixel point, as the reference color value; determine the color value of the non-center pixel point in the selected area And the difference value of the reference color value; if the difference value is greater than the preset difference threshold, it is determined that the corresponding non-center pixel point is different from the center pixel point color; otherwise, it is determined that the corresponding non-center pixel point is the same as the center pixel point; decomposition final determination Different cases of color, determine the color category.

[0090] Further, the color category identifier can be determined according to the difference threshold interval belonging to the difference value; wherein the difference threshold interval can be set by the skilled person, or adjusting the reference color value in accordance with the center pixel point. Specifically, if the difference threshold interval falling in different difference values is different, the color category of the corresponding non-central pixel point is characterized. Among them, there is no overlap between different threshold intervals.

[0091] It will be appreciated that the color category is determined by introducing a central pixel point corresponding to the reference color value, and the color value of the color value and the reference color value of the reference color value are performed by the difference between the color value corresponding to the color value of the color value by the pixel point. The color conditions are quantitatively described, providing data support for the determination of the candidate zone, enriches the method of determining the color category. It should be noted that the color category is determined by the above manner, and only the specific category identification of different color categories is required, and the amount of data operation is reduced.

[0092] Optionally, determine the pixel boundaries of each category color, which may be combined with the same color category, to obtain each color category correspondence color area; a pixel boundary of each color area as a corresponding color class.

[0093] Alternatively, it is possible to determine the pixel boundaries of each category color, which identifies a pixel point different from the peripheral color category, as a boundary pixel point; combination of the boundary pixel point of the same color category to obtain the pixel boundary.

[0094] In an alternative embodiment, according to the pixel boundaries of each category color, the candidate area can be: directly overlapping the larger boundary (e.g., maximum boundary) formed by the pixel boundary, as a candidate zone boundary; defining the candidate area boundary Region, as a candidate area; or determine the external geometry of the area defined by the candidate area boundary, the external geometry corresponds to the area as the candidate area. Wherein, the external set graphics may be rectangular or circular, and the present disclosure is not specifically defined.

[0095] It will be appreciated that the determination of the candidate zone is performed by the above technical solution, and the determination process is convenient and fast, and the operation is small.

[0096] In another alternative embodiment, according to the pixel boundaries of each category color, the candidate zone may be: the area defined by the pixel boundary of each category color as a reference area; the area defined according to the pixel boundary of different color categories The area and reference area determine the candidate area.

[0097] Among them, the reference area is superimposed in the largest region obtained by superposition of different color categories, for the determination foundation of the candidate area. It should be noted that the reference area is the aforementioned alternative embodiment, and the maximum boundary defined by the pixel boundary is superimposed.

[0098] Exemplary, the regional area, crop reference area, and reference area defined according to the pixel boundary of different color categories, determines whether the candidate area is determined: the area area of the region defined according to the pixel boundary of the different color categories; According to the base area, the reference area is cropped, the candidate area is obtained.

[0099] Optionally, according to each color category corresponds to the area area, the basic area, such as a smaller (eg, minimum) area area in each area area, the determined mean area area, the selected median area area, etc. As the base area. Corresponding, according to the base area, the crop parameter is determined; where the crop parameters may include crop length and / or crop width; according to the crop parameters, the reference area is cut according to the corresponding area boundary according to the base area, and the reference area is cropped, and it includes the smaller area area. Regional candidate area.

[0100] In order to avoid excessive cases in the candidate area, optionally, when cropping the reference area according to the base area, the corresponding area is corresponding to the regional boundary, and the corresponding area is expanded, and the corresponding area is expanded. The boundary of the region; cropped the reference area based on the adjusted area boundary to obtain a candidate area including the base area corresponding to the area.

[0101] In a specific implementation, it is also possible to define the same or similar to the different color categories in the candidate zone, such as the difference between the corresponding area area of the different color categories, so that each color category in the candidate area tends to be uniform.

[0102] It will be appreciated that by introducing the regional area of the region defined by the region of the different color categories, the crop reference area is cropped, and the reference area is too large, the enlargement is poor, and the rationality of the selected candidate area is improved. Increased the rationality of the local amplification image, it is convenient to select the pixel point selection from the local amplification image, reducing the occurrence of the pixel point corresponding color misleading, thereby improving the accuracy of the color identification result.

[0103] Exemplary, according to the area area of the region defined according to the pixel boundary of the different color categories, the candidate area may be: the base area is determined according to the area area of the region defined according to the pixel boundary of the different color categories; determines the base area corresponding to the base area External geometry of the area; translation of the external geometry along the distribution of different color categories, to obtain a candidate area. Wherein, the external geometry may be a rectangle, or a positive polygon, or the like, the disclosure does not limit the shape of the external geometry. In a concrete implementation, the selection of external binding graphics can be performed according to the distribution of different color categories.

[0104] The present disclosure is refined to the regional area and reference zone defined by the pixel boundary of each category color class, according to the area area and reference area defined by the pixel boundaries of the different color categories. Region. The above technical solution defines a method of determining the candidate area, and the determination of the candidate zone is performed by introducing a regional area and reference area of different color classes, and perfecting the method of determining the candidate zone is provided for the determination of the local amplified image.

[0105] S303, amplify the corresponding area image of the candidate area to obtain a partially enlarged image.

[0106] S304, in response to the pixel point selection operation of the local amplified image, identify the target color of the selected pixel point, and output the target color.

[0107] The present disclosure is refined to determine the candidate zone to determine the candidate area, and improve the determination method of the candidate area according to the color properties of the candidate zone. At the same time, the determination of the candidate area based on the color properties is dynamically performed, and the determined candidate area is improved, and the accurate selection of the image to be identified is provided, which provides a guarantee for the accurate selection of the subsequent pixel point, thereby helping to avoid the misunderstanding of the pixel point color. The occurrence of the situation, thereby helping to improve the accuracy of color identification results.

[0108] Based on the technical solutions of the above embodiments, the present disclosure also provides a preferred embodiment of a color recognition method.

[0109] See Figure 4A A color identification method shown, including:

[0110] S401, in response to the photo operation or photo selection operation, top the identification image.

[0111] See Figure 4B The image is logged to be identified.

[0112] S402 determines the selected area in response to the regional selection operation.

[0113] Where the area selection operation can be a click operation.

[0114] S403, determine if the pixel point color center pixel point in the selected area is similar to the non-center pixel point color value; if yes, S404A is executed; otherwise, S404B is executed.

[0115] S404A, using the first extension, expanded in the selected area, to obtain the area to be displayed. Continue S405.

[0116] S404B, using the second expansion multiple, expanded in the selected area to obtain the area to be displayed. Continue to execute S405.

[0117] Among them, the second spreading multiple is greater than the first extension.

[0118] S405, the identification image is enlarged, resulting in a global amplified image, and showing a global enlarged image to be displayed to be displayed to be partially enlarged image.

[0119] See Figure 4C and Figure 4D The selected area is a full-red area in the image to be identified, and the partially enlarged image of the obtained area to be displayed, and the selected area is "red, black, white" handover area in the image to be identified, and the resulting display Compared with the local amplification of the area, the latter corresponds to the extension of the area to be displayed larger, that is, the number of different color blocks corresponding to the latter.

[0120] S406, in response to the transform operation of the local amplified image, update the local enlarged image. Wherein, the transform operation includes at least one of a zoom operation, a mobile operation, and a restore operation.

[0121] The zoom operation can refer to the operation, the movement operation can be a drag operation, the restore operation can be long pressing, etc., the present disclosure does not limit the specific mode of operation of each operation, only to ensure that each operation can distinguish.

[0122] S407, in response to the pixel point selection operation of the local amplified image, identify the target color of the selected pixel point.

[0123] S408, text show target color, and / or voice output target color.

[0124] Through the techniques of the present disclosure, it is possible to assist special people in color identification without introducing hardware costs. In addition, the above technical solution is convenient and suitable for various people, and has a wide range of advice. At the same time, by showing local enlarged images, it is easy to perform pixel point selection, avoiding the occurrence of color misleading in color miscibility, and improves color identification result accuracy.

[0125] As an implementation of the above color identification methods, the present disclosure also provides an alternative embodiment of an execution apparatus for performing a color recognition method. See Figure 5 One color recognition device 500, including: to be identified image display module 501, a candidate area determining module 502, an enlarged display module 503, and a color identification display module 504. in,

[0126] The image display module 501 is used to display the image to be identified;

[0127] The candidate zone determination module 502 is configured to determine the candidate area according to the selected area in response to the area selection operation of the image to be identified.

[0128] The amplification display module 503 is used to enlarge the corresponding area image of the candidate zone to obtain a partially enlarged image;

[0129] The color identification display module 504 is configured to identify the target color of the selected pixel point in response to the pixel point selection operation of the local amplified image, and output the target color.

[0130] The present disclosure is shown to be identified by the image display module to be identified; the candidate display area is determined according to the selected area by the candidate zone determination module, according to the selected area, according to the selected area; enlargement of the candidate area corresponding area Image, obtain a local enlarged image; the target color of the selected pixel point is identified by the color identification display module in response to the pixel point selection operation of the local amplified image, and outputs the target color. With the above technical solution, it is possible to provide color identification help to a population that cannot distinguish at least some of the color, such as a color blind, or the weak, so that the class can make color identification without the need for a color blind correction device, reducing hardware cost, and capable Adapted to different categories of user bases, wide adaptation range. At the same time, the present disclosure is enlarged by introducing a candidate area, and is enlarged to the candidate zone corresponding to the region image, which is obtained to provide a local amplified image selected by the pixel point, which is convenient for the operator to accurately perform the pixel point selection, avoiding the deviation due to the pixel point selection. The target color, the actual color does not match the actual color of the expectation, improves the accuracy of color recognition results.

[0131] In an alternative embodiment, the candidate zone determines module 502, including:

[0132] The candidate area determining unit is used to determine the candidate area according to the color attribute of the selected area.

[0133] In an alternative embodiment, the candidate zone determines units, including:

[0134] Color category identification subunits, used to identify color categories in the selected area, and determine the pixel boundaries of each category color;

[0135] The candidate area determines the subunit to determine the candidate zone based on the pixel boundaries of each category color.

[0136] In an alternative embodiment, the color category identifies subunit, including:

[0137] The first color category identifies the slave unit for identifying whether or not the preset color category is included in the selected area, and determines the color category according to the identification result; or,

[0138] The second color category identifies the slave unit for the color value of the central pixel point of the selected area as the reference color value, and determines according to the color value of the non-center pixel point in the selected area and the difference value of the reference color value. Color category.

[0139] In an alternative embodiment, the candidate zone determines subunits, including:

[0140] The reference area determines the slave unit, used to limit the area defined by the pixel boundaries of each category as a reference area;

[0141] The candidate zone determines from the unit to determine the candidate area in accordance with the area area of the region and the reference area defined according to the pixel boundary of the different color categories.

[0142] In an alternative embodiment, the candidate area determines from the unit, including:

[0143] The basic area determines the sub-unit for determining the regional area of the region defined according to the pixel boundary of different color categories;

[0144] The reference area is cut from the unit, used to crop the reference area according to the base area, to obtain the candidate zone, so that the different color category corresponding to the different color category in the candidate area is smaller than the set threshold.

[0145] In an alternative embodiment, the magnifying display module 503, comprising:

[0146] The global amplification image obtains a unit for an enlargement of the image to be identified to obtain a global amplified image;

[0147] The local amplification image display unit is configured to show the candidate area corresponding to the candidate area in the global amplified image to obtain the local amplified image.

[0148] In an alternative embodiment, the global amplification image obtains a unit, including:

[0149] The target amplification ratio determination subunit is used to determine the target amplification ratio according to each category color in the candidate zone;

[0150] The overall enlargement image is obtained from the subunit, and is used to enlarge the image to be enlarged according to the target amplification ratio, resulting in the global amplifying image.

[0151] In an alternative embodiment, the magnifying display module 503, comprising:

[0152] The target amplification ratio determination subunit is used to determine the target amplification ratio according to each category color in the candidate zone;

[0153] The partially enlarged image obtains a sub-unit for an enlargement of the candidate zone corresponding to the area image to obtain the local amplification image according to the target amplification ratio.

[0154] In an alternative embodiment, the target amplification ratio determines subunits, including:

[0155] The reference amplification ratio determines the slave unit for determining the reference amplification ratio according to the ratio of the candidate area and the local amplification display area;

[0156] The target amplification ratio determines the slave unit for selecting the target amplification ratio from the candidate zone in accordance with the reference amplification ratio, selecting the target amplification ratio.

[0157] The color recognition device described above can perform the color identification method provided by any of the present disclosure, including functional modules and beneficial effects corresponding to each color identification method.

[0158] In the technical solution of the present disclosure, the processing, storage, use, processing, transmission, transmission, providing and disclosure of the image to be identified, and is in line with the relevant laws and regulations, and does not violate the publicquence prison.

[0159] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0160] Image 6 A schematic block diagram showing an example electronic device 600 that can be used to implement the embodiments of the present disclosure is shown. Electronic equipment is intended to represent various forms of digital computers, such as laptop, desktop computer, workbench, personal digital assistant, server, blade server, large computer, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connection and relationship, and their functions are merely examples, and are not intended to limit the implementation of the present disclosure as described herein.

[0161] like Image 6 As shown, device 600 includes a calculation unit 601, which can perform various appropriate appropriate Action and processing. In the RAM 603, the various programs and data required for device 600 can also be stored. Computing unit 601, ROM 602, and RAM 603 are connected to each other through bus 604. The input / output (I / O) interface 605 is also connected to the bus 604.

[0162]The plurality of components in the device 600 are connected to the I / O interface 605, including: input unit 606, such as a keyboard, a mouse, or the like, the output unit 607, such as various types of displays, speakers, etc .; storage unit 608, such as a disk, CD, etc. ; And communication unit 609, such as a network card, modem, wireless communication transceiver, and the like. The communication unit 609 allows the device 600 to exchange information / data exchange information / data from a computer network and / or a variety of telecommunications networks such as the Internet.

[0163] Computing unit 601 can be a common and / or dedicated processing component having processing and computing power. Some examples of the calculation unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various operational machine learning model algorithms calculation unit, digital signal processing (DSP), and any suitable processor, controller, microcontroller, etc. The calculation unit 601 performs the respective methods and processing described above, such as a color identification method. For example, in some embodiments, the color identification method can be implemented as a computer software program that is tangily included in a machine readable medium, such as a storage unit 608. In some embodiments, the partial or all of the computer program may be loaded and / or mounted on the device 600 via the ROM 602 and / or communication unit 609. When the computer program is loaded into the RAM 603 and executed by the calculation unit 601, one or more steps of the color identification method described above can be performed. Alternatively, in other embodiments, calculating unit 601 can be configured to perform color identification methods by other suitable means (e.g., by means of firmware).

[0164] Various embodiments of the systems and techniques described above in this article can be in digital electronic circuitry, integrated circuitry, field programmable gate array (FPGA), dedicated integrated circuit (ASIC), special standard product (ASSP), chip system System (SOC), load programmable logic (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include: implementing in one or more computer programs, which can be performed and / or interpreted on a programmable system including at least one programmable processor, the programmable processor Can be a dedicated or universal programmable processor, data and instructions can be received from the storage system, at least one input device, and at least one output device, and transmit data and instruction to the storage system, the at least one input device, and at least one An output device.

[0165] The program code for implementing the method of the present disclosure can be written any combination of one or more programming languages. These program code can provide a processor or controller for a general purpose computer, a dedicated computer, or another programmable data processing device such that the program code is performed by the processor or the controller to perform the functions specified in the flowchart and / or block diagram / The operation is implemented. The program code can be performed entirely on the machine, partially executed on the machine, execute on the machine as a stand-alone software package and is performed on the remote machine or on the remote machine or server.

[0166] In the context of the present disclosure, the machine readable medium may be a tangible medium, which may contain or store procedures for instruction execution systems, devices or devices, or combined with instruction execution systems, devices, or devices. The machine readable medium can be a machine readable signal medium or a machine readable storage medium. Machine readable media can include, but are not limited to, electron, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any suitable combination of the above. More specific examples of machine readable storage media include electrical connection, portable computer disc, hard disk, random access memory (RAM), read-only memory (ROM) based on one or more lines of electrical connection, read-only memory (ROM), erased-programmable read-only memory (EPROM or flash memory), fiber optic, convenient compact disk read only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above.

[0167] In order to provide interaction with the user, the systems and techniques described herein can be implemented on a computer, which is: display device for displaying information to the user (eg, a CRT (cathode ray tube) or LCD (liquid crystal display) monitor ); And keyboards and pointing devices (e.g., mouse or trackballs), users can provide input to the computer by this keyboard and the pointing device. Other types of devices can also be used to provide interactions with the user; for example, feedback to the user can be any form of sensing feedback (eg, visual feedback, audible feedback, or haptic feedback); and can be in any form (including Acoustic input, voice input, or haptic input) to receive input from the user.

[0168] The systems and techniques described herein can be implemented in a computing system (e.g., a data server) including the background component, or a computing system (e.g., application server) including an intermediate member component, or a computing system including a front end member (eg, With a user computer with a graphical user interface or a web browser, the user can interact with the system and technique of the system and technology described herein by this graphical user interface or the network browser), or including this background component, an intermediate member, Or in any combination of the front end member. The components of the system can be connected to each other by digital data communication (eg, a communication network) in any form or a medium. Examples of the communication network include: LAN, WAN (WAN), and the Internet.

[0169] Computer systems can include clients and servers. Client and servers are generally away from each other and are usually interacting through a communication network. The relationship between the client and the server is generated by running on the corresponding computer and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, a host product in the cloud computing service system to solve the traditional physical host and VPS service, the existence management is high, and the business extensibility is weak. defect. The server can also be a server of a distributed system, or a server that combines block chains.

[0170] Artificial intelligence is the study that makes the computer to simulate some of the human thinking process and intelligent behavior (such as learning, reasoning, thinking, planning, etc.), both hardware levels of technology also have software level technologies. Artificial intelligence hardware technology generally includes such as sensors, special artificial intelligent chips, cloud computing, distributed storage, big data processing, etc., artificial intelligence software technology mainly includes computer visual technology, speech recognition technology, natural language processing technology and machine learning / depth Learning technology, big data processing technology, knowledge map technology, etc.

[0171] Cloud computing refers to a shared physical or virtual resource pool that access elastic extensible through the network. Resources can include servers, operating systems, networks, software, applications, and storage devices, etc., and can serve on demand. The way the resources are deployed and managed by resources. Through cloud computing technology, it can be used for artificial intelligence, block chains, etc., model training provides efficient and powerful data processing capabilities.

[0172] It should be understood that the various forms of forms shown above can be used, reordering, increasing, and deleting steps. For example, the steps described in the present disclosure can be performed in parallel, and may be performed sequentially, as long as the technical solutions disclosed in the present disclosure, this paper is not limited thereto.

[0173] DETAILED DESCRIPTION OF THE INVENTION The scope of the present disclosure is not constituted. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and replacements can be made according to design requirements and other factors. Any modification, equivalent replacement and improvement, etc. within the spirit and principles of the present disclosure, should be included within the scope of the present disclosure.

PUM

Description & Claims & Application Information

We can also present the details of the Description, Claims and Application information to help users get a comprehensive understanding of the technical details of the patent, such as background art, summary of invention, brief description of drawings, description of embodiments, and other original content. On the other hand, users can also determine the specific scope of protection of the technology through the list of claims; as well as understand the changes in the life cycle of the technology with the presentation of the patent timeline. Login to view more.

Classification and recommendation of technical efficacy words

improve accuracy

Color recognition method and device, equipment and storage medium

A color recognition and color classification technology, applied in the field of image processing, can solve the problem of color blindness or color weakness unable to distinguish natural spectral colors, etc., to achieve the effect of improving accuracy

AI-Extracted Technical Summary

Problems solved by technology

Method used

Abstract

Image

Examples

Example Embodiment

PUM

Description & Claims & Application Information

Similar technology patents

Techniques for sentiment analysis of data using a convolutional neural network and a co-occurrence network

Adaptive fault detection method for airplane rotation actuator driving device based on deep learning

Portable reference station for local differential GPS corrections

Automatic or semi-automatic cooking equipment and batch charging mechanism thereof

Communication terminal apparatus and communication system

Classification and recommendation of technical efficacy words

Golf club head with adjustable vibration-absorbing capacity

Stent delivery system with securement and deployment accuracy

Method for improving an HS-DSCH transport format allocation

Catheter systems

Gaming Machine And Gaming System Using Chips