CN111127333B - Improved color enhancement method for two-color vision-oriented color image - Google Patents
Improved color enhancement method for two-color vision-oriented color image Download PDFInfo
- Publication number
- CN111127333B CN111127333B CN201911103035.7A CN201911103035A CN111127333B CN 111127333 B CN111127333 B CN 111127333B CN 201911103035 A CN201911103035 A CN 201911103035A CN 111127333 B CN111127333 B CN 111127333B
- Authority
- CN
- China
- Prior art keywords
- color
- image
- dominant
- improved
- vision
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 239000003086 colorant Substances 0.000 claims abstract description 35
- 230000004438 eyesight Effects 0.000 claims abstract description 20
- 230000002708 enhancing effect Effects 0.000 claims abstract description 4
- 201000004569 Blindness Diseases 0.000 claims description 19
- 230000000007 visual effect Effects 0.000 abstract description 12
- 238000012937 correction Methods 0.000 abstract description 4
- 238000004364 calculation method Methods 0.000 description 7
- 238000006243 chemical reaction Methods 0.000 description 6
- 206010008795 Chromatopsia Diseases 0.000 description 5
- 208000006992 Color Vision Defects Diseases 0.000 description 5
- 201000007254 color blindness Diseases 0.000 description 5
- 230000004075 alteration Effects 0.000 description 4
- 201000000761 achromatopsia Diseases 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000001415 gene therapy Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 208000036693 Color-vision disease Diseases 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 210000003484 anatomy Anatomy 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000004456 color vision Effects 0.000 description 1
- 230000006735 deficit Effects 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007935 neutral effect Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/90—Dynamic range modification of images or parts thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Image Processing (AREA)
Abstract
The invention discloses an improved method for enhancing the color of a color image facing to two-color vision. The invention acquires the main colors of the image by clustering in the u 'v' space for the image to be subjected to color correction, and then converts u 'and v' into R and theta. Then, by remapping the theta of the several dominant colors to theta' according to the order of confusion priority, the improved method is followed, and for each pixel point belonging to the kth dominant color, the original angle theta is calculated according to the original angle theta 0 Correspondingly adjust to theta 0 +(θ k ‑θ′ k ) Thereby ensuring the difference of hues between different colors. Meanwhile, R based on the pixel point ij 、L ij Mean value of dominant color radius
And a maximum dominant color radius difference R max ‑R min The brightness of the image is adjusted, so that the visual difference among colors is improved, the visual experience of a dichromatic viewer is improved, and the dependence on clustering results is reduced.
Description
Technical Field
The invention relates to an improved method for enhancing the color of a color image facing to two-color vision. Belongs to the technical fields of computational vision, digital image processing, color enhancement and the like.
Background
Modern anatomy verifies that three cone cells exist in the human eye, namely, an L cone cell (sensitive to long-band light), an M cone cell (sensitive to middle-band light) and an S cone cell (sensitive to short-band light). The so-called "achromatopsia" and "color weakness" are caused by the lack or damage of certain cone cells, and cannot normally distinguish the visible light of a specific wave band in the natural spectrum, so that other means can be radically cured except for the gene therapy with huge cost.
Among the most types of color vision disorders, the most common types are dichroism and abnormal trichromatism, that is, the inconvenience of color discrimination caused by the lack or damage of one type of cone cells, the ratio of which exceeds 95%, and can be classified into red blindness (protanaria, lack of L cone cells), green blindness (deuteranopa, lack of M cone cells), blue blindness (tritanaria, lack of S cone cells) according to the type of lack of cone cells, and the ratio of male to female is 7% globally and the ratio of female to 0.5% in terms of red blindness and green blindness. The data report of public welfare organization ColorBlind Awareness established in the united kingdom for color vision impairment shows that 4.5% of the population worldwide is currently achromatopsia. At present, the China has more than 6000 ten thousand people with dyschromatopsia, but under the current medical level, the cost of gene therapy is huge, the gene therapy is affordable by unusual families, and the dyschromatopsia brings inconvenience to the dyschromatopsia people only in the color differentiation of daily life, but has no direct threat on health and safety, so that the method has an important role in improving the visual experience of the dyschromatopsia people when the digital image is displayed.
In 1997, hans bretetel et al in Computerized simulation of color appearance for dichromats, provided a specific conversion calculation method by converting linear algebra to obtain a visual simulation matrix from trichromatism to dichroism based on the assumption that neutral colors have no change in the dichroism visual angle in the LMS color space, according to the fact that the color blindness of red, green and blue can correctly distinguish 475nm light from 575nm light and the color blindness of blue can correctly distinguish 485nm light from 660nm light. In 1999 the number of years,
vienot et al, in Digital Video Colourmaps for Checking the Legibility of Displays by Dichromats, gives a specific transformation matrix in the LMS color space.
Since many expert students have conducted rich discussions and researches on the image enhancement method for two-color vision, mainly there are remapping, brightness adjustment and other methods, but because of the lack of two-color vision cone cells, three-dimensional color information is accepted as two-dimensional information, and the trouble of manufacturing new color confusion is often caused when remapping and brightness adjustment are not combined with image content information, so that the current research is mainly focused on color enhancement based on image content.
Disclosure of Invention
The invention mainly provides an improved image color enhancement method facing to two-color vision based on image content. For an image to be subjected to color correction, clustering is carried out in a u 'v' space to obtain the main colors of the image, and then u 'and v' are converted into R and theta. Then, by remapping the theta of the several dominant colors to theta' according to the order of confusion priority, the improved method is followed, and for each pixel point belonging to the kth dominant color, the original angle theta is calculated according to the original angle theta 0 Correspondingly adjust to theta 0 +(θ k -θ′ k ) Thereby ensuring the difference of hues between different colors. Meanwhile, R and dominant color radius mean value based on the pixel point
And a maximum dominant color radius difference R max -R min The brightness of the image is adjusted, so that the visual difference among colors is improved, the visual experience of a dichromatic viewer is improved, and the dependence on clustering results is reduced.
The technical scheme adopted by the invention for solving the practical application problem is a processing method for enhancing the image color of a dichromatic visual crowd, and the processing process is in a Lu 'v' color space, and the specific steps are as follows:
step (1) converts the image from the sRGB color space to the Lu 'v' color space, and clusters in the u 'v' space, obtaining image dominant color information based on the image content, wherein the conversion from the sRGB color space to the XYZ color space is calculated as follows:
wherein the method comprises the steps of
The conversion of the XYZ color space to the Lu 'v' color space is calculated as follows:
step (2) the (u ', v') of the dominant color obtained by the clustering is expressed as (u 'in the u' v 'space' con ，v′ con ) = (0.678,0.501) as center (here, lack of L cone cells results in two-color vision, and lack of M cone cells results in (u' con ，v′ con ) = (-1.217,0.782) as center, if S cone cells are absent, (u '' con， v′ con ) = (0.257,0.0) as a center), converting (u ', v') into polar representation (R, θ), the conversion method is as follows:
step (3) angle θ for the dominant color with highest priority of current adjacent confusion without remapping k Remapping to θ 'according to the improved remapping method' k Until all dominant colors have been remapped, the calculation formula for the adjacent confusion priority is as follows:
M k ＝(θ k -θ′ k-1 ) 2 +(θ′ k+1 -θ k ) 2 ，k＝1，2..K
wherein if the kth dominant color has not been remapped, θ' k In the absence of θ' k By theta k Instead of calculating the value of θ' 0 And θ' K+1 For the angle approximation of the two boundaries of the u ' v ' space, θ ' 0 ＝-2.169782992，θ′ K+1 = -1.46003987, selecting among the main colors that have not been mappedThe nearest confusion priority is taken to be the largest (i.e. M in the primary color is not remapped) i The largest dominant color).
The two-color vision lacking L-cone cells or M-cone cells was remapped according to the following improved method:
wherein, for a dichromatic viewer lacking L cone cells, θ d ＝-1.49939917，θ′ 0 And theta'. K+1 The value is unchanged, a is an hyper-parameter of an algorithm, the adjustment angle can be controlled within a certain range, so that the problem that the image naturalness is reduced due to overlarge image change before and after enhancement is avoided, and the attempt a epsilon [0.1,0.15 ] is tried]The naturalness of the image can be better maintained, and the image enhancement is better affected; for the dichromatic vision lacking M cone cells, θ d ＝1.64282159285，θ′ 0 ＝1.73292258701，θ′ K+1 ＝1.99216810406，a∈[0.0365，0.0548]The value of a should be different depending on the application scenario.
The two-color vision lacking S-cone cells was remapped as follows:
wherein, θ' 0 ＝-0.476983507，θ′ K+1 ＝-0.23049605；a∈[0.0347，0.0521]The value of a should be different depending on the application scenario.
Step (4) if the pixel point (i, j) is affiliated to the kth class dominant color cluster, then the theta is calculated ij Go and theta k The same adjustment is carried out to obtain theta' ij ＝θ ij +(θ′ k -θ k ) (k=1, 2,..k), where K is the dominant color class cluster number and K is the total number of clustered dominant colors;
step (5) for each pixel point (i, j), based on R ij Mean value of dominant color radius R
And dominant color maximum radius difference |R max -R min The brightness is adjusted to obtain L':
wherein b is an hyper-parameter of the algorithm, and through an attempt, when b epsilon [20, 30], the difference of pixels with different hues in brightness can be improved under the condition of better keeping the naturalness of the image, so that the contrast of mixed colors under the two-color vision is improved, and the value of b is different according to the application scene in specific application.
Step (6) converts the enhanced (R, θ ') to (u ', v ') by:
in combination with L ', the (L ', u ", v") is converted back to (R ', G ', B ') for display.
And obtaining the color image after color enhancement.
The technical scheme provided by the invention has the beneficial effects that:
the adjustment sequence before remapping of the angle theta is improved, the color enhancement effect is improved, and when the Euclidean distance between different main colors in the u 'v' plane after enhancement is longer; the improvement of remapping is carried out on a dichromatic vision person (red blindness) lacking L cone cells, a dichromatic vision person (green blindness) lacking M cone cells and a dichromatic vision person (blue blindness) lacking S cone cells, and the distribution balance of the enhanced colors on a u 'v' plane is ensured while the color contrast of the image is enhanced; further utilize R ij Mean value of dominant color radius R
And dominant color maximum radius difference |R max -R min Adjusting brightness for colors with the same angle θ but different RThe brightness of the images is enhanced (namely, the colors which can be confused by the two-color vision person) and the difference of original confusion colors of the finally obtained images in eyes of the dyschromatopsia person becomes obvious, so that the visual effect is improved, and meanwhile, the dependence on clustering results is reduced.
Drawings
FIG. 1 is an original view of a normal three-color viewing angle embodiment;
fig. 2 is a primary color (cluster centroid of c=2 and c=4) of the red blind confusing color at the trichromatic viewing angle in fig. 1;
FIG. 3 is an illustration of an embodiment at a red blind simulated viewing angle;
fig. 4 is a confusing dominant color at the red blind viewing angle in fig. 3 (cluster centroids of c=2 and c=4, i.e. the red blind simulated viewing angle of fig. 2);
fig. 5 is a graph of clustering results obtained after clustering an original image by Kmeans (k=4), wherein different clusters are represented by different gray levels;
FIG. 6 is a diagram of the original color enhancement results for a normal three-color viewing angle embodiment;
fig. 7 is an enhanced dominant color (cluster centroid of c=2 and c=4) of the red blind confusing color at the trichromatic viewing angle of fig. 6;
FIG. 8 is a diagram of the color enhancement results of the original image of the embodiment at a red blind simulated viewing angle;
fig. 9 is a confusing dominant color at the red blind viewing angle in fig. 8 (cluster centroids of c=2 and c=4, i.e. the red blind simulated viewing angle of fig. 6);
fig. 10 is a robustness check result diagram.
Fig. 11 is a flow chart of the present invention.
Detailed Description
The technical scheme of the invention can automatically carry out the flow by adopting the computer software technology. For a better understanding of the technical solution of the present invention, the following describes the present invention in further detail with reference to the drawings and examples. An embodiment of the invention is a real image of red blind and difficult to distinguish colors. Referring to fig. 1 and 11, the flow of the embodiment of the present invention includes the following steps:
step (1) clustering the image colors in a u 'v' space;
step (2) converting (u ', v') of several dominant colors obtained by clustering into (R, θ), finding the degree of confusion M that is not remapped and is currently in proximity to k Angle θ of maximum dominant color k Remapping to θ 'following the improved method' i ；
Step (3) when the angle θ of the dominant color is still maintained k Repeating step (2) when the remapping is not performed;
step (4) for each pixel point in the image, θ ' = (θ ') is performed on it ' i -θ i ) Adjustment of +θ, wherein the color of the pixel belongs to the dominant color of the i-th class;
step (5) utilizes R ij 、L ij Mean value of dominant color radius R
And dominant color maximum radius difference |R max -R min Adjusting brightness;
step (6) converts the image from the Lu 'v' color space to the sRGB color space.
Step (1) converts the image from the RGB color space to the XYZ color space and then to the Lu 'v' color space, where dominant color clustering is performed, here exemplified by Kmeans clustering.
For an image to be enhanced, it is first transformed into XYZ color space, as follows:
wherein the method comprises the steps of
The conversion of the XYZ color space to the Lu 'v' color space is calculated as follows:
the Kmeans clustering algorithm pseudocode is as follows:
the Kmeans clustering result (k=4) is shown in fig. 2, where gray values 0, 50, 100, 150 represent the cluster categories 1,2, 3, 4, respectively.
In the step (2), converting (u ', v') of several dominant colors obtained by clustering into (R, theta), and according to the current proximity confusion degree M k Angle θ of maximum dominant color k Remapping to θ 'following the improved method' k The specific procedure of (2) is as follows.
The cluster centers of the four main colors obtained in the step (1) are shown in the following table:
| clustering categories | u′ k | v′ k |
| 1 | 0.204268 | 0.493217 |
| 2 | 0.358061 | 0.512938 |
| 3 | 0.220898 | 0.522419 |
| 4 | 0.187399 | 0.524124 |
Since it is image enhancement for the dichromatic viewer (red blindness) lacking L-cone cells, (u' con ，v′ con ) = (0.678,0.501), the (u ', v') of the 4 dominant colors are converted to (R, θ) according to the following formula
The results are shown in the following table:
| clustering categories | R k | θ k |
| 1 | 0.47379636 | -1.58722 |
| 2 | 0.32016166 | -1.5335 |
| 3 | 0.45760337 | -1.52397 |
| 4 | 0.491146 | -1.5237 |
The proximity confusion M of the currently un-remapped dominant color is calculated according to the following calculation formula i ：
M k ＝(θ k -θ′ k-1 ) 2 +(θ′ k+1 -θ k ) 2 ，k＝1，2..K
Wherein, when facing to the people lacking L cone cell dichroism vision, θ' 0 ＝-1.46003987，θ′ K+1 ＝-2.169782992。
| Clustering categories | θ k | M k |
| 1 | -1.58722 | 0.342261974 |
| 2 | -1.5335 | 0.002976842 |
| 3 | -1.52397 | 9.08785E-05 |
| 4 | -1.5237 | 0.004052372 |
At this time, the proximity confusion M of the 1 st color 1 Maximum, first for the angle θ of the first type of color 1 By adjusting the colors with large adjacent confusion degree preferentially, more angle space can be reserved for the colors which are adjusted subsequently, and an angle mapping formula for the two-color vision person lacking the L cone cells is shown as follows, namely theta d ＝-1.49939917：
Solving to obtain theta' 1 ＝-1.6872。
Wherein θ d Intersection of color confusion lines (u 'for over-red blindness (lack of L-cone cytodichroism)' con ，v′ con ) = (0.678,0.501) perpendicular to the straight line (v '=13.86 u' -2.273) fitted to the color range visible in the u 'v' plane of the red blind
Is (u '=0.2026, v' = 0.5353) (i.e. v '=13.86 u' -2.273 and +.>
Is a cross point of (2); for green blindness, the solution process only changes (u' con ，v′ con ) = (-1.217,0.782) the straight line fitted in the color range visible on the u 'v' plane is the same as the red blindness, and the specific process of solving the above-mentioned solution of the red blindness is not specifically given here, θ for the green blindness d = 1.642821592848115; and according to Hans Brettel and +.>
Vihenot simulates a blue blind dichromatic viewing angle, the range of colors visible in the u 'v' plane is not straight, so another mapping method is used), where a takes 0.1.
θ 1 Mapped to theta' 1 Continuing to calculate the adjacent confusion M for the currently unmapped color k 。
At this time, the proximity confusion M of the class 2 color 2 Maximum, remapping is θ' 2 = -1.6063, continuing to calculate the adjacent confusion M for the currently unmapped color i 。
At this time, the proximity confusion M of the 3 rd color 3 Maximum, remapping is θ' 3 = -1.5651, finally remapped class 4 colors that have not been remapped, θ' 4 = -1.5126. The results before and after final remapping are shown below:
| clustering categories | θ k | θ k | θ′ k -θ k |
| 1 | -1.5872 | -1.6872 | -0.1 |
| 2 | -1.5335 | -1.6063 | -0.0728 |
| 3 | -1.52397 | -1.5651 | -0.04113 |
| 4 | -1.5237 | -1.5126 | 0.0111 |
Step (4) for the pixel point at the image (i, j), θ 'is performed' ij ＝(θ′ k -θ k )+θ ij Wherein the color of the pixel belongs to the dominant color of the kth class;
step (5) for each pixel point (i, j), based on R ij Mean value of dominant color radius R
And dominant color maximum radius difference |R max -R min And adjusting the brightness to obtain L', wherein the calculation formula is as follows: />
Wherein b is taken here as 25,
|R max -R min meter of IThe calculation is as follows:
step (6) converts the image from the Lu 'v' color space to the sRGB color space.
And converting the remapped theta ' into u ' v ', wherein the conversion method is as follows:
the Lu 'v' color space is converted to XYZ color space as follows:
the XYZ color space is converted to the RGB color space by the following method:
wherein c=r, G, B.
And obtaining the final color correction image.
The feasibility of the technical scheme of the invention is proved as follows:
the chromatic aberration is a widely applied color similarity calculation method at present, L * a * b * The chromatic aberration is the currently mainstream and most easy-to-calculate chromatic aberration calculation mode, defined as:
wherein DeltaL * ，Δa * ，Δb * The difference of three channels of the two colors in Lab space.
In the original image, the colors (red and green) originally mixed are clustered by Kmeans in two clusters of c=2 and c=4, as shown in the following table of cluster centroid RGB values (cluster RGB average):
converting it from RGB to L * a * b * The results are shown below:
calculating according to a color difference formula to obtain a color difference result as follows:
color confusing L in example graph * a * b * Chromatic aberration
| Original image | The method enhances the color difference | |
| Normal viewing angle of three colors | 59.07364448 | 54.21596997 |
| View angle of achromatopsia | 11.03078338 | 36.75849249 |
The smaller the color difference is, the higher the similarity between the two colors is, and the color difference improvement percentage under the red blind view angle after the color enhancement of the example image is calculated to be 233.24% by taking the red blind view angle as an example, meanwhile, the color discrimination is obviously improved under the simulated red blind view angle by comparing the attached figures 3 and 8 through subjective judgment, and the visual experience of the dichroism vision is effectively improved.
From the color difference and visual experience obtained by experiments, the color correction method provided by the invention can better improve the identification degree of different colors after the image is enhanced, and the method can effectively improve the visual experience of a dichromatic viewer.
Robustness test
The method has lower dependence on the clustering result of the adopted clustering method and the clustering quantity obtained during clustering, namely, the method can still have better performance under the conditions of non-ideal clustering result and non-ideal clustering quantity, and the result is shown in figure 10, so that the method can be proved to have higher robustness and stability.
The foregoing is a further detailed description of the invention in connection with the preferred embodiments, and is not intended to limit the practice of the invention to such description. It will be understood by those skilled in the art that various changes in detail may be effected therein without departing from the scope of the invention as defined by the claims appended hereto.
Claims (6)
1. An improved method for enhancing the color of a color image facing to two-color vision is characterized by comprising the following steps:
step (1) clustering the image colors in a u 'v' space;
step (2) converting (u ', v') of several dominant colors obtained by clustering into(R, θ) find the current proximity confusion M without remapping k Angle θ of maximum dominant color k Remapping to θ 'following the improved method' i ；
Step (3) when the angle θ of the dominant color is still maintained k Repeating step (2) when the remapping is not performed;
step (4) for each pixel point in the image, θ ' = (θ ') is performed on it ' k -θ k ) Adjustment of +θ, wherein the color of the pixel belongs to the dominant color of the i-th class;
step (5) utilizes R ij 、L ij Mean value of dominant color radius R
And dominant color maximum radius difference |R max -R min Adjusting brightness;
step (6) converts the image from the Lu 'v' color space to the sRGB color space.
2. An improved two-color vision-oriented color image color enhancement method according to claim 1, characterized in that said step (2) is performed by a proximity confusion priority M for the dominant color of the image to be remapped i To confirm the adjustment sequence of the dominant color angle, the specific implementation is as follows:
M k ＝(θ k -θ' k-1 ) 2 +(θ' k+1 -θ k ) 2 ,k＝1,2..K。
3. an improved two-color vision-oriented color image color enhancement method according to claim 1 or 2, characterized in that said angle θ for remapping to be performed in step (2) i The method is concretely realized as follows:
for both the dichroism (red blindness) lacking L cone cells and the dichroism (green blindness) lacking M cone cells, remapping was performed according to the improved method, specifically as follows:
for dichroism (red blindness) lacking L cone cells: θ d ＝-1.49939917，θ′ 0 ＝-2.169782992，θ′ K+1 ＝-1.46003987；
For dichroism (green blindness) lacking M cone cells: θ d ＝1.64282159285，θ′ 0 ＝1.73292258701，θ′ K+1 ＝1.99216810406。
4. An improved colour enhancement method for colour images facing dichroism according to claim 1 or 2, characterized in that in step (2) the dichroism lacking S-cone cells (blue blindness) is remapped as follows:
wherein, θ' 0 ＝-0.476983507，θ' K+1 ＝-0.23049605；
The value of a depends on the type of the oriented dichromatic vision and the application scene.
5. An improved two-color vision-oriented color image color enhancement method as defined in claim 3, wherein step (4) is specifically implemented as follows:
based on the principal color of the clustered image, R is used for each pixel point (i, j) ij Mean value of dominant color radius R
And dominant color maximum radius difference |R max -R min The luminance is adjusted to obtain adjusted luminance L ', and the luminance L' is solved as follows: />
The value of b in a specific application is different according to the application scene.
6. An improved two-color vision-oriented color image color enhancement method as defined in claim 5, wherein step (6) is specifically implemented as follows:
the enhanced (R, θ') is converted to (u ", v") by the following formula:
and (L ', u ', v ') is converted back to (R ', G ', B ') for display in combination with the luminance L '.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911103035.7A CN111127333B (en) | 2019-11-12 | 2019-11-12 | Improved color enhancement method for two-color vision-oriented color image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911103035.7A CN111127333B (en) | 2019-11-12 | 2019-11-12 | Improved color enhancement method for two-color vision-oriented color image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111127333A CN111127333A (en) | 2020-05-08 |
| CN111127333B true CN111127333B (en) | 2023-05-05 |
Family
ID=70495238
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911103035.7A Active CN111127333B (en) | 2019-11-12 | 2019-11-12 | Improved color enhancement method for two-color vision-oriented color image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111127333B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114863095B (en) * | 2022-03-25 | 2023-11-28 | 电子科技大学 | Answer sheet image segmentation method based on color conversion |
| CN117116176A (en) * | 2022-11-01 | 2023-11-24 | 深圳市Tcl云创科技有限公司 | Content display method, apparatus, electronic device, and computer-readable storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006086882A (en) * | 2004-09-16 | 2006-03-30 | Ricoh Co Ltd | Image forming apparatus, image forming method, program, and recording medium |
| CN101853492A (en) * | 2010-05-05 | 2010-10-06 | 浙江理工大学 | Method for fusing night-viewing twilight image and infrared image |
| CN106249406A (en) * | 2016-08-30 | 2016-12-21 | 喻阳 | Improve Color perception and correct artificial intelligence's lens and the method for designing of achromatopsia color weakness vision |
| CN106530250A (en) * | 2016-11-07 | 2017-03-22 | 湖南源信光电科技有限公司 | Low illumination color image enhancement method based on improved Retinex |
| CN109710791A (en) * | 2018-12-14 | 2019-05-03 | 中南大学 | A kind of multi-source color image color moving method based on significant filter |
| US10331207B1 (en) * | 2013-03-15 | 2019-06-25 | John Castle Simmons | Light management for image and data control |
| CN110322520A (en) * | 2019-07-04 | 2019-10-11 | 厦门美图之家科技有限公司 | Image key color extraction method, apparatus, electronic equipment and storage medium |
-
2019
- 2019-11-12 CN CN201911103035.7A patent/CN111127333B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006086882A (en) * | 2004-09-16 | 2006-03-30 | Ricoh Co Ltd | Image forming apparatus, image forming method, program, and recording medium |
| CN101853492A (en) * | 2010-05-05 | 2010-10-06 | 浙江理工大学 | Method for fusing night-viewing twilight image and infrared image |
| US10331207B1 (en) * | 2013-03-15 | 2019-06-25 | John Castle Simmons | Light management for image and data control |
| CN106249406A (en) * | 2016-08-30 | 2016-12-21 | 喻阳 | Improve Color perception and correct artificial intelligence's lens and the method for designing of achromatopsia color weakness vision |
| CN106530250A (en) * | 2016-11-07 | 2017-03-22 | 湖南源信光电科技有限公司 | Low illumination color image enhancement method based on improved Retinex |
| CN109710791A (en) * | 2018-12-14 | 2019-05-03 | 中南大学 | A kind of multi-source color image color moving method based on significant filter |
| CN110322520A (en) * | 2019-07-04 | 2019-10-11 | 厦门美图之家科技有限公司 | Image key color extraction method, apparatus, electronic equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| Metamer Mismatching and Its Consequences for Predicting How Colours are Affected by the Illuminant;Xiandou Zhang etal;《2015 IEEE International Conference on Computer Vision Workshop (ICCVW)》;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111127333A (en) | 2020-05-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110276727B (en) | Color vision disorder-oriented color image color enhancement method | |
| CN111127333B (en) | Improved color enhancement method for two-color vision-oriented color image | |
| US7873213B2 (en) | Systems and methods for color-deficient image enhancement | |
| CN105184754B (en) | Method for enhancing picture contrast | |
| CN106997584A (en) | A kind of haze weather image enchancing method | |
| WO2018113050A1 (en) | Drive method and drive apparatus of display panel | |
| CN111915508B (en) | Image texture detail enhancement method for color vision disorder | |
| WO2019056445A1 (en) | Color gamut mapping method and device | |
| CN108711142A (en) | Image processing method and image processing apparatus | |
| Walraven et al. | Color displays for the color blind | |
| CN107092753A (en) | A kind of intelligent city's planning system | |
| CN110175969A (en) | Image processing method and image processing apparatus | |
| CN111105383B (en) | Three-color vision-oriented image fusion color enhancement method | |
| CN107122613A (en) | A kind of DICOM image display methods based on pseudo-colours | |
| CN115345788B (en) | Method and device for improving image color contrast under vision of person with color vision abnormality | |
| CN105208362B (en) | Image colour cast auto-correction method based on gray balance principle | |
| CN106599185B (en) | HSV-based image similarity identification method | |
| EP2802139B1 (en) | Image color adjusting method and electronic device using the same | |
| TWI438718B (en) | Image processing method and system by using adaptive inverse hyperbolic curve | |
| CN101036606B (en) | Method for rectifying the daltonism on the basis of self-adapted mapping | |
| CN110880306B (en) | Color restoration correction method for medical display | |
| CN104796679A (en) | Laser display color adjusting method and device thereof | |
| CN110767150B (en) | Method for solving squint color cast problem of LED screen | |
| CN107154015B (en) | Color blindness correction method based on regional mapping | |
| CN108564632B (en) | Black and white image colorizing algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Shen Xuming Inventor after: Feng Jianhua Inventor after: Zhang Xiandou Inventor after: Wang Yong Inventor before: Zhang Xiandou Inventor before: Shen Xuming Inventor before: Feng Jianhua |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |