CN107845080B - Card image enhancement method - Google Patents
Card image enhancement method Download PDFInfo
- Publication number
- CN107845080B CN107845080B CN201711191889.6A CN201711191889A CN107845080B CN 107845080 B CN107845080 B CN 107845080B CN 201711191889 A CN201711191889 A CN 201711191889A CN 107845080 B CN107845080 B CN 107845080B
- Authority
- CN
- China
- Prior art keywords
- image
- card image
- card
- highlight
- space
- 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 43
- 238000012937 correction Methods 0.000 claims abstract description 11
- 238000012015 optical character recognition Methods 0.000 claims abstract description 10
- 238000012545 processing Methods 0.000 claims description 14
- 238000006243 chemical reaction Methods 0.000 claims description 11
- 238000001514 detection method Methods 0.000 claims description 7
- 230000000877 morphologic effect Effects 0.000 claims description 6
- 238000005260 corrosion Methods 0.000 claims description 3
- 230000007797 corrosion Effects 0.000 claims description 3
- 238000003708 edge detection Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000001131 transforming effect 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
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/04—Context-preserving transformations, e.g. by using an importance map
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
- G06T5/30—Erosion or dilatation, e.g. thinning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration using histogram techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/80—Geometric correction
-
- 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
- G06T5/94—Dynamic range modification of images or parts thereof based on local image properties, e.g. for local contrast enhancement
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Character Input (AREA)
Abstract
The invention provides a card image enhancement method, which comprises the following steps: obtaining a card image acquired by a mobile terminal; denoising the acquired card image to obtain a first card image; converting the first card image from an RGB space to a YCrCb space, extracting a Y space image in the first card image to judge whether the first card image has highlight or not; if the first card image has a highlight area, removing the highlight area by converting the brightness characteristic of the highlight area to obtain a second card image; performing inclination correction on the second card image to obtain a third card image; and carrying out optical character recognition on the third card image, and outputting a recognition result. Determining highlight of the card image from the extracted Y-space image by converting the card image from an RGB space to a YCrCb space, and removing the highlight by converting a luminance characteristic of a highlight region; therefore, highlight and inclination problems of the card image are quickly removed and corrected, so that the recognition rate of subsequent optical character recognition is improved.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a card image enhancement method for mobile terminal acquisition.
Background
In people's daily life, it is often necessary to obtain relevant information from certificates. Generally, an image of a card such as a certificate is captured by an imaging device such as a scanner, then the image is subjected to some necessary processing by using an image processing technology, and finally, related information in the card image is identified and extracted by using an Optical Character Recognition (OCR) technology. In recent years, with the development of mobile communication and internet technologies, a large number of mobile terminal devices such as smart phones and tablet computers are gradually becoming an indispensable part of people's lives. These mobile devices all have a camera function, and are widely used in related industries such as finance and internet, and meanwhile, new changes and challenges are brought to traditional card image processing and recognition. For the card image collected by the mobile terminal, two common problems are that the image has high light and inclination, which makes the relevant information in the card image possibly unable to be identified or the identification result is wrong, and finally the relevant information input may fail. Therefore, there is a need for fast removal and correction of high light and tilt of card images captured by a moving end.
Highlight detection and removal of a single image has been a difficult point. The main methods at present are as follows: the first method is complementary color, and fills the image area by extracting the surrounding image features to remove highlights, which has the disadvantages of large amount of information to be processed, slow processing speed, and loss of the original image features in the highlights area. The second method is to remove the high light according to the reflection model, which has the disadvantage that the repair results are poor if the emission model is too simple; if the reflection model is too complex, the processing speed is slow and the algorithm is less applicable. The third method is illumination constraint color compensation, which combines the common color compensation method with the illumination constraint condition to remove the highlight, and has the disadvantages of complex algorithm, large information amount of processing and slow processing speed. In summary, the conventional method solves the problems of excessive calculation amount and slow processing speed of highlight.
Disclosure of Invention
The invention provides a card image enhancement method, which solves the problems in the prior art.
In order to solve the above problem, an embodiment of the present invention provides a card image enhancement method, including the following steps:
obtaining a card image acquired by a mobile terminal;
denoising the acquired card image to obtain a first card image;
converting the first card image from an RGB space to a YCrCb space, extracting a Y space image in the first card image to judge whether the first card image has highlight or not;
if the first card image has a highlight area, removing the highlight area by converting the brightness characteristic of the highlight area to obtain a second card image;
performing inclination correction on the second card image to obtain a third card image;
and carrying out optical character recognition on the third card image, and outputting a recognition result.
As an embodiment, the transforming the first card image from RGB space to YCrCb space, extracting Y space image therein to determine whether the first card image has highlights includes the following steps:
converting the first card image from an RGB space to a YCrCb space, and extracting a Y space image in the first card image as an image to be detected;
setting a highlight detection threshold value, and setting a pixel value lower than the highlight detection threshold value in an image to be detected as 0;
carrying out binarization processing on an image to be detected, and then carrying out morphological operation to obtain a highlight area of a first card image, wherein the morphological operation comprises opening operation and corrosion;
and confirming whether the first card image is highlight or not according to the corroded image.
As an embodiment, if the first card image has a highlight area, the highlight area is removed by converting the brightness characteristic of the highlight area to obtain the second card image, specifically including the following steps:
normalizing the brightness Y value in the Y space image to obtain a normalized Y image;
performing histogram equalization operation on the normalized Y image to obtain an equalized Y image;
and replacing the original brightness Y value with the new brightness Y value through a conversion polynomial according to the image brightness conversion criterion on the equalized Y image to obtain a second card image.
In one embodiment, the normalization process of the luminance Y value in the Y space image is performed such that the maximum value of the luminance Y value is mapped to 1 and the minimum value of the luminance Y value is mapped to 0.
As an embodiment, the conversion polynomial is:
y＝2.5208Y5-4.8316Y4+2.3675Y3-0.4013Y2+0.9998Y。
as an embodiment, the tilt correcting the second card image to obtain the third card image specifically includes the following steps:
graying the second card image;
then, edge detection is carried out by using a Canny operator to obtain a second card binary image;
calculating the slope of the edge straight line of the second card binary image by using a least square method;
and calculating according to the slope to obtain an angle, and then rotating the image to obtain a third card image.
As an implementation mode, a median filtering method is adopted to carry out denoising processing on the acquired card image.
Compared with the prior art, the invention has the beneficial effects that: determining whether the card image has highlight according to the extracted Y space image by converting the card image collected by the mobile terminal from an RGB space to a YCrCb space; if the highlight exists, removing the highlight by converting the brightness characteristic of the highlight area, and then performing inclination correction; therefore, highlight and inclination problems of the card image are quickly removed and corrected, so that the recognition rate of subsequent optical character recognition is improved.
Drawings
FIG. 1 is a flow chart of a card image enhancement method of the present invention;
FIG. 2 is a flowchart illustrating steps 300 of a card image enhancement method according to the present invention;
FIG. 3 is a flowchart illustrating the steps 400 of the card image enhancement method according to the present invention;
FIG. 4 is a flowchart illustrating steps 500 of a card image enhancement method according to the present invention;
FIG. 5 is an original card image captured by the mobile terminal;
FIG. 6 is a highlight region of a first card image processed by the card image enhancement method of the present invention;
FIG. 7 is a second card image obtained after highlight removal by the card image enhancement method of the present invention;
FIG. 8 is a third card image obtained after the card image enhancement method of the present invention is tilt-corrected.
Detailed Description
The above and further features and advantages of the present invention will be apparent from the following, complete description of the invention, taken in conjunction with the accompanying drawings, wherein the described embodiments are merely some, but not all embodiments of the invention.
As shown in fig. 1, a card image enhancement method includes the following steps:
s100: obtaining a card image collected by the mobile terminal, as shown in fig. 5;
s200: denoising the acquired card image to obtain a first card image, and in the embodiment, denoising the acquired card image by using a median filtering method;
s300: converting the first card image from an RGB space to a YCrCb space, extracting a Y space image in the first card image to judge whether the first card image has highlight or not;
s400: if the first card image has a highlight area, removing the highlight area by converting the brightness characteristic of the highlight area to obtain a second card image;
s500: performing inclination correction on the second card image to obtain a third card image;
s600: and performing Optical Character Recognition (OCR) on the third card image, and outputting a recognition result.
As shown in fig. 2, step S300 specifically includes the following steps:
s301: converting the first card image from an RGB space to a YCrCb space, and extracting a Y space image in the first card image as an image to be detected;
s302: setting a highlight detection threshold value, and setting a pixel value lower than the highlight detection threshold value in an image to be detected as 0;
s303: performing binarization processing on an image to be detected, and then performing morphological operations to obtain a highlight area of the first card image, as shown in fig. 6, in the present embodiment, the morphological operations include opening operation and corrosion;
s304: and confirming whether the first card image is highlight or not according to the corroded image.
As shown in fig. 3, step S400 specifically includes the following contents:
s401: normalizing the brightness Y value in the Y space image to obtain a normalized Y image, wherein in the embodiment, the normalization is to map the maximum value of the brightness Y value to 1 and map the minimum value of the brightness Y value to 0;
s402: performing histogram equalization operation on the normalized Y image to obtain an equalized Y image;
s403: for the equalized Y image, replacing the original luminance Y value with the new luminance Y value by a conversion polynomial according to the image luminance transformation criterion to obtain a second card image, as shown in fig. 7, in this embodiment, the conversion polynomial is: Y2.5208Y5-4.8316Y4+2.3675Y3-0.4013Y2+0.9998Y, the image brightness conversion criterion is to reduce the brightness of the high light area in the image and ensure the brightness of the area with lower image brightness to be basically unchanged, therefore, the invention comprehensively considers the image brightness conversion criterion and the calculation amount and adopts the fifth-order polynomial to carry out conversion.
The traditional image tilt correction method mainly comprises a Fourier transform-based method, a Hough transform-based method, a projection-based method, a cross correlation-based method, a K-nearest neighbor cluster method and the like. The main disadvantages of these methods are the excessive amount of computation and the poor adaptability of some algorithms.
In this embodiment, as shown in fig. 4, the tilt correction step S500 specifically includes the following steps:
s501: graying the second card image;
s502: then, edge detection is carried out by using a Canny operator to obtain a second card binary image;
s503: calculating the slope of the edge straight line of the second card binary image by using a least square method;
s504: the angle is calculated from the slope and the image is then rotated to obtain a third card image, as shown in fig. 8.
Compared with the traditional image inclination correction method, the inclination correction method adopted by the invention can quickly correct the inclined card image, and has the advantages of simple method, small calculated amount and easy realization.
The card image collected by the mobile terminal is converted into a YCrCb space from an RGB space, and whether the card image has high light or not is determined according to the extracted Y space image; if the highlight exists, removing the highlight by converting the brightness characteristic of the highlight area; therefore, highlight and inclination problems of the card image are quickly removed and corrected, so that the recognition rate of subsequent optical character recognition is improved.
The above-mentioned embodiments are provided to further explain the objects, technical solutions and advantages of the present invention in detail, and it should be understood that the above-mentioned embodiments are only examples of the present invention and are not intended to limit the scope of the present invention. It should be understood that any modifications, equivalents, improvements and the like, which come within the spirit and principle of the invention, may occur to those skilled in the art and are intended to be included within the scope of the invention.
Claims (5)
1. A card image enhancement method is characterized by comprising the following steps:
obtaining a card image acquired by a mobile terminal;
denoising the acquired card image to obtain a first card image;
converting the first card image from an RGB space to a YCrCb space, extracting a Y space image in the first card image to judge whether the first card image has highlight or not;
if the first card image has a highlight area, removing the highlight area by converting the brightness characteristic of the highlight area to obtain a second card image;
performing inclination correction on the second card image to obtain a third card image;
carrying out optical character recognition on the third card image, and outputting a recognition result;
the method for judging whether the first card image has highlight or not by converting the first card image from an RGB space to a YCrCb space and extracting a Y space image in the first card image specifically comprises the following steps:
converting the first card image from an RGB space to a YCrCb space, and extracting a Y space image in the first card image as an image to be detected;
setting a highlight detection threshold value, and setting a pixel value lower than the highlight detection threshold value in an image to be detected as 0;
carrying out binarization processing on an image to be detected, and then carrying out morphological operation to obtain a highlight area of a first card image, wherein the morphological operation comprises opening operation and corrosion;
confirming whether the first card image is highlight or not according to the corroded image;
if the first card image has a highlight area, removing the highlight area by converting the brightness characteristic of the highlight area to obtain a second card image, specifically comprising the following steps of:
normalizing the brightness Y value in the Y space image to obtain a normalized Y image;
performing histogram equalization operation on the normalized Y image to obtain an equalized Y image;
and replacing the original brightness Y value with the new brightness Y value through a conversion polynomial according to the image brightness conversion criterion on the equalized Y image to obtain a second card image.
2. The card image enhancement method according to claim 1, wherein the normalization processing of the luminance Y value in the Y space image is performed such that the maximum value of the luminance Y value is mapped to 1 and the minimum value of the luminance Y value is mapped to 0.
3. The card image enhancement method of claim 1, wherein the conversion polynomial is:
y＝2.5208Y5-4.8316Y4+2.3675Y3-0.4013Y2+0.9998Y。
4. the card image enhancement method according to any one of claims 1 to 3, wherein the tilt correction is performed on the second card image to obtain a third card image, and the method specifically comprises the following steps:
graying the second card image;
then, edge detection is carried out by using a Canny operator to obtain a second card binary image;
calculating the slope of the edge straight line of the second card binary image by using a least square method;
and calculating according to the slope to obtain an angle, and then rotating the image to obtain a third card image.
5. The card image enhancement method of claim 1, wherein the acquired card image is denoised by a median filtering method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711191889.6A CN107845080B (en) | 2017-11-24 | 2017-11-24 | Card image enhancement method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711191889.6A CN107845080B (en) | 2017-11-24 | 2017-11-24 | Card image enhancement method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107845080A CN107845080A (en) | 2018-03-27 |
| CN107845080B true CN107845080B (en) | 2020-03-20 |
Family
ID=61679350
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711191889.6A Active CN107845080B (en) | 2017-11-24 | 2017-11-24 | Card image enhancement method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107845080B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111275648B (en) * | 2020-01-21 | 2024-02-09 | 腾讯科技（深圳）有限公司 | Face image processing method, device, equipment and computer readable storage medium |
| CN115578695B (en) * | 2022-11-21 | 2023-03-28 | 昆明理工大学 | Water gauge water level machine vision detection method and device with free shooting visual angle |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20140054690A (en) * | 2012-10-29 | 2014-05-09 | 서울여자대학교 산학협력단 | Automatic fusion system and the method of t2-weighted mr image and diffusion weighted mr image |
| CN104616268A (en) * | 2015-02-17 | 2015-05-13 | 天津大学 | Underwater image restoration method based on turbulence model |
| KR20150144727A (en) * | 2014-06-17 | 2015-12-28 | 주식회사 유진로봇 | Apparatus for recognizing location mobile robot using edge based refinement and method thereof |
| CN105205792A (en) * | 2015-09-18 | 2015-12-30 | 天津大学 | Underwater image enhancement method based on brightness and chrominance separation |
| US9330340B1 (en) * | 2014-05-06 | 2016-05-03 | Google Inc. | Noise estimation for images using polynomial relationship for pixel values of image features |
| CN105957042A (en) * | 2016-06-07 | 2016-09-21 | 北京理工大学 | Highlight region eliminating method of endoscopic image |
| CN107274385A (en) * | 2017-05-18 | 2017-10-20 | 常州大学 | Textile flaw detection method based on class grid pattern and its Gabor characteristic |
-
2017
- 2017-11-24 CN CN201711191889.6A patent/CN107845080B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20140054690A (en) * | 2012-10-29 | 2014-05-09 | 서울여자대학교 산학협력단 | Automatic fusion system and the method of t2-weighted mr image and diffusion weighted mr image |
| US9330340B1 (en) * | 2014-05-06 | 2016-05-03 | Google Inc. | Noise estimation for images using polynomial relationship for pixel values of image features |
| KR20150144727A (en) * | 2014-06-17 | 2015-12-28 | 주식회사 유진로봇 | Apparatus for recognizing location mobile robot using edge based refinement and method thereof |
| CN104616268A (en) * | 2015-02-17 | 2015-05-13 | 天津大学 | Underwater image restoration method based on turbulence model |
| CN105205792A (en) * | 2015-09-18 | 2015-12-30 | 天津大学 | Underwater image enhancement method based on brightness and chrominance separation |
| CN105957042A (en) * | 2016-06-07 | 2016-09-21 | 北京理工大学 | Highlight region eliminating method of endoscopic image |
| CN107274385A (en) * | 2017-05-18 | 2017-10-20 | 常州大学 | Textile flaw detection method based on class grid pattern and its Gabor characteristic |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107845080A (en) | 2018-03-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9898808B1 (en) | Systems and methods for removing defects from images | |
| EP1910994B1 (en) | Binarization of an image | |
| US8611662B2 (en) | Text detection using multi-layer connected components with histograms | |
| US9679354B2 (en) | Duplicate check image resolution | |
| CN109993161B (en) | Text image rotation correction method and system | |
| US9245194B2 (en) | Efficient line detection method | |
| CN108965646B (en) | Image processing apparatus, image processing method, and program | |
| CN108146093A (en) | A kind of method for removing bill seal | |
| CN107845080B (en) | Card image enhancement method | |
| CN114820594A (en) | Method for detecting edge sealing defects of plate based on image, related equipment and storage medium | |
| WO2024016632A1 (en) | Bright spot location method, bright spot location apparatus, electronic device and storage medium | |
| CN110738604A (en) | Canny operator and Hough transformation-based large-dip angle certificate image correction method and system | |
| JP2014174803A (en) | Moire removal method for pattern recognition, moire removal device using the same, and program | |
| Lin et al. | License plate recognition based on mathematical morphology and template matching | |
| CN112419207A (en) | Image correction method, device and system | |
| JP2015115067A (en) | Method and device for extracting distorted straight line from image | |
| JP5870745B2 (en) | Image processing apparatus, binarization threshold value calculation method, and computer program | |
| CN105809677B (en) | Image edge detection method and system based on bilateral filter | |
| CN111445402A (en) | Image denoising method and device | |
| CN115862044A (en) | Method, apparatus, and medium for extracting target document part from image | |
| JP6661268B2 (en) | Image processing apparatus, image processing method, and program | |
| WO2022056875A1 (en) | Method and apparatus for segmenting nameplate image, and computer-readable storage medium | |
| CN114219760A (en) | Reading identification method and device of instrument and electronic equipment | |
| JP2020149184A (en) | Information processor and control method thereof and program | |
| JP2019021085A (en) | Image processing program, image processing method, and image processing device |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder |
Address after: Xinyada technology building, 3888 Jiangnan Avenue, Binjiang District, Hangzhou City, Zhejiang Province 310000 Patentee after: Sinyada Technology Co.,Ltd. Address before: Xinyada technology building, 3888 Jiangnan Avenue, Binjiang District, Hangzhou City, Zhejiang Province 310000 Patentee before: SUNYARD SYSTEM ENGINEERING Co.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder |