CN107169466B - Palm print image quality comprehensive evaluation method based on rank-sum ratio method - Google Patents
Palm print image quality comprehensive evaluation method based on rank-sum ratio method Download PDFInfo
- Publication number
- CN107169466B CN107169466B CN201710379075.9A CN201710379075A CN107169466B CN 107169466 B CN107169466 B CN 107169466B CN 201710379075 A CN201710379075 A CN 201710379075A CN 107169466 B CN107169466 B CN 107169466B
- Authority
- CN
- China
- Prior art keywords
- palm print
- print image
- palm
- block
- rank
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/12—Fingerprints or palmprints
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30168—Image quality inspection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Collating Specific Patterns (AREA)
Abstract
The invention relates to a palm print image quality comprehensive evaluation method based on a rank-sum ratio method, which comprises the following steps: step S01: primarily evaluating the quality of the palm print image, and determining that the palm print image which cannot meet the primary evaluation requirement is unqualified in comprehensive evaluation of the quality of the palm print image; step S02: cutting and numbering the palm print image to form a palm print image block evaluation sequence; step S03: respectively calculating each evaluation index of each palm print image block; step S04: carrying out comprehensive quality evaluation on all the palm print image blocks by adopting a rank-sum ratio method; step S041: ranking each evaluation index of the palm print image block; step S042: defining a rank and ratio matrix of all palm print image blocks; step S043: sequentially calculating the rank sum ratio of each palmprint block; step S05: presetting a plurality of rank and ratio thresholds, counting the number of the palm print image blocks larger than different thresholds, and determining the quality level of each palm print image. The invention can rapidly and accurately complete the comprehensive evaluation of the palm print image quality through the palm print image quality preliminary evaluation and the palm print image quality comprehensive evaluation two-stage evaluation mode based on the rank-sum ratio method.
Description
Technical Field
The invention provides a palm print image quality comprehensive evaluation method based on a rank-sum ratio method, belongs to the field of biological feature recognition, and particularly relates to a palm print recognition method.
Background
Compared with other biological feature recognition technologies, the palm print automatic recognition technology has extremely good accuracy and high economic practicability, and the current application prospect is very bright. Palm print image quality evaluation is one of the most direct and important factors influencing palm print identification accuracy. The high-quality palm print image can ensure that the palm print recognition system extracts more palm print characteristic data, and can make the palm print compare algorithm distinguish all palm print data in the palm print database more efficiently. Although the palm print image and the fingerprint image are similar in texture and form, and the common quality evaluation method is also similar to the fingerprint image quality evaluation method, the traditional fingerprint image quality evaluation method cannot be directly carried out to the palm print image quality evaluation problem because: (1) the area of the palm print image is many times larger than that of the fingerprint image, and if the palm print is evaluated in a fingerprint image quality evaluation mode, the palm print evaluation can be finished in a long time; (2) because the palm physiological structure is different from that of the fingers, the pressing force of the palm is more uneven and the force points are more dispersed, so that the difference among all areas of the palm print image is large, and the image quality is not suitable to be evaluated by adopting a global quality unified evaluation mode of the whole image; (3) when a plurality of evaluation indexes are used for palm print image quality evaluation, a strategy for refining the indexes and providing a comprehensive and scientific evaluation result is lacked.
The prior palm print image quality comprehensive evaluation method has no effective solution and successful application in the three aspects.
Disclosure of Invention
The invention solves the problems: aiming at the problem of evaluating the quality of the palm print image, the method overcomes the defect that the evaluation time is long for the prior art; uneven pressing force degree and dispersed acting points; the palm print image quality comprehensive evaluation method based on the rank-sum ratio method can quickly and accurately complete the comprehensive evaluation of palm print image quality through the palm print image quality preliminary evaluation and the palm print image quality comprehensive evaluation two-stage evaluation mode based on the rank-sum ratio method.
The overall process of the invention is shown in figure 1 and comprises the following steps:
step S01: primarily evaluating the quality of the palm print image, and determining that the palm print image which cannot meet the primary evaluation requirement is unqualified in comprehensive evaluation of the quality of the palm print image;
wherein said preliminary evaluation generally comprises the following quality evaluation: whether the palm print image is complete and the size of the palm blank area are two evaluation indexes. The evaluation index is characterized by simple calculation method and quick calculation.
Wherein, the evaluation indexes are specifically explained as follows: if the height of the palm print image is H pixels and the width is W pixels, then (1) whether the palm print image is complete. The evaluation can be made as a ratio of the palm print area to the whole image: cpalm＝Ipalm/(H.times.W), wherein IpalmFor pixels in the palm print region of the imageThe number of the cells; (2) size of palm space. In consideration of the convenience in designing the threshold, the index adopts the reciprocal of the number of pixels in the palm space area as an evaluation index. The calculation formula is Bpalm＝1/Iblank，IblankThe number of pixels in the palm space area.
The palm center blank area refers to an area where the palm center is not collected by the palm print image collecting device. The reason for this is that the palm center is slightly depressed toward the back of the hand in the physiological structure compared with the peripheral outline of the palm, so that the palm center is not collected by the palm print collecting device. The area identified by triangle ABC in fig. 2 is a palm space area. In comprehensive evaluation of palm print image quality, the smaller the area of the region, the better.
If any one of the preliminary evaluation indexes does not exceed a preset threshold value, the comprehensive evaluation of the palm print image quality is considered to be unqualified, and the subsequent comprehensive evaluation of the palm print image quality is not performed.
Step S02: cutting and numbering the palm print image into blocks to form a palm print image block quality evaluation sequence;
the palm print image block cutting means that the palm print image is cut into a plurality of image blocks according to a certain size, and the purpose is to split the palm print image into the plurality of palm print image blocks, so that each evaluation index of each palm print image block can be conveniently calculated in the subsequent step of S03 in a parallel manner. Assuming the height of each palm print image block as HblockPixel of width WblockThe number of image blocks which can be cut out of the palm print image is
Wherein H and W can be independently selected from HblockAnd WblockAnd (4) trimming.
Step S03: respectively calculating each evaluation index of each palm print image block;
wherein, the evaluation indexes generally comprise: (1) the palm print area occupies the image block; (2) the dryness and the humidity of the image block; (3) the image block definition degree; (4) degree of consistency of the palm print line direction. These evaluation indexes evaluate the palm print image quality from different angles.
The method for calculating the ratio of the palm print area in the image block comprises the following steps: cblock＝Iblock/(Hblock×Wblock)，IblockThe number of pixels in the palm print area in the image block.
Wherein the dryness and the humidity of the image block are defined as DRYblock＝|128-Eblock|，EblockIs the expected value of the gray values of all pixels in the image block. DRYblockThe higher the palm size, the more dry or wet the palm was during collection.
Wherein, the image block definition can be the standard deviation S of all pixel gray values in the palm print image blockblockEvaluation was carried out.
The degree of the consistency of the palm print line direction can be obtained by filtering the image block by using a Gabol filtering method in image processing. In the present invention, the calculation result is represented by CONblockAnd (4) showing.
Step S04: carrying out comprehensive quality evaluation on all the palm print image blocks by adopting a rank-sum ratio method;
wherein, the quality comprehensive evaluation of all the palm print image blocks by adopting the rank-sum ratio method is further divided into the following substeps:
step S041: ranking each evaluation index of the palm print image block;
step S042: defining a rank and ratio matrix of all palm print image blocks;
step S043: and sequentially calculating the rank and ratio of each palm print image block.
Wherein, the evaluation indexes in step S041 are ranked, and the palm print area in (1) of the palm print image block is generally occupied by the image block; (2) the dryness and the humidity of the image block; (3) the image block definition degree; (4) and 4, the consistency degree of the directions of the lines is obtained, and the four indexes are ranked. Wherein, the three indexes of (1), (3) and (4) are high priority and (2) are low priority.
Wherein, the definition in step S042 is fullThe method comprises the following specific steps of: defining a rank and ratio matrix as
A total of N block images participate in the quality comprehensive evaluation, and 4 represents 4 evaluation indexes described in step S041. Any element in the matrix is RijAnd representing the rank ordering result of the j index of the ith palm print image block.
In step S043, the rank and ratio of each palm print image block are sequentially calculated, and the specific method includes: for any palm print image block i, the rank sum ratio is as follows:
step S05: presetting a plurality of rank and ratio thresholds, counting the number of the palm print image blocks larger than different thresholds, and determining the quality level of each palm print image;
wherein, the preset plurality of rank sum ratio thresholds are implemented by setting a rank sum ratio threshold Y1And Y2Wherein Y is1Value of greater than Y2The value of (c).
Counting the number of the palm print image blocks larger than different thresholds, wherein the specific method is to sequentially and respectively count the RSR of each palm print image blockiA value of Y or more1And Y2Are respectively z1And z2Wherein is greater than or equal to Y1RSR ofiThe value is also simultaneously greater than Y2。
Wherein, the quality grade of each palm print image is determined by setting the threshold value of comprehensive evaluation of palm print image quality as "excellent" as Z1The threshold value for comprehensively evaluating the palm print image quality as "good" is Z2. The judgment rule is as follows: if z is1≥Z1Finally, the comprehensive evaluation result of the palm print image quality is determined to be excellent; if z is1<Z1And z is2≥Z2Finally, the comprehensive evaluation result of the palm print image quality is determined to be good; if the two judgment conditions are not met, the comprehensive evaluation of the palm print image quality is finally determinedThe result was "pass".
The invention has the following beneficial technical effects:
(1) the comprehensive evaluation method for the quality of the palm print image based on the rank-sum ratio method can preliminarily evaluate the quality of the palm print image and screen out the palm print image with unqualified quality; and (5) cutting the palm print image into blocks, and calculating each quality evaluation index of each image block in a parallel mode. The method reduces the evaluation time, and simultaneously avoids the problems that the standard is difficult to control caused by uneven palm pressing strength and dispersed acting points on image quality evaluation; the rank-sum ratio method can scientifically and effectively revive various palm print image evaluation indexes and provide a comprehensive evaluation result.
(2) The method of the invention provides a strategy for condensing each index and providing a comprehensive and scientific evaluation result when evaluating the quality of the palm print image by aiming at a plurality of indexes. With the increasingly wide application of palm print identification, the beneficial technical effects of the invention are more obvious;
(3) the method for cutting the palm print and calculating various palm print quality evaluation indexes in parallel effectively overcomes the defect of long palm print image quality evaluation calculation time on one hand; on the other hand, the defects that the differences among all areas of the palm print image are large and the global quality of the whole image is not suitable for unified evaluation due to uneven pressing force and dispersed force points during palm acquisition are overcome.
Drawings
FIG. 1 is a flow chart of a method of practicing the present invention;
fig. 2 is a schematic view of a palm-center blank area.
Detailed Description
The present invention will be described in detail below with reference to the accompanying drawings and examples.
As shown in fig. 1, the method comprises the steps of:
step S01: primarily evaluating the quality of the palm print image, and determining that the palm print image which cannot meet the primary evaluation requirement is unqualified in comprehensive evaluation of the quality of the palm print image;
for a given acquired palm print image,assume that the image height H is 2304 pixels and the width W is 2304 pixels. The following indicators were initially calculated: (1) whether the palm print image is complete. Cutting the image into a palm print area and a background area by utilizing the difference of the gray values of all pixels in the palm print area and the background area and adopting a related method of an image processing technology, and counting the number I of pixel points in the palm print areapalm4034396, thus Cpalm4034396/(2304 × 2304) ═ 0.76; (2) the reciprocal of the pixel number of the palm blank area. And randomly sampling pixel points at the center of the palm print area stripped from the image. If the gray value of the pixel point is close to the gray value of the image background area, the palm blank area is gradually found by adopting methods such as seed filling in computer graphics and the like. Counting the number of pixel points in the palm blank area as Ipalm4290, then Bpalm＝1/Iblank1/4290. Setting CpalmHas a threshold value of 0.5, BpalmWhen the preliminary evaluation of the palm print image quality comprehensive evaluation is completed, 1/9000, the palm print image quality comprehensive evaluation is rejected, and the process proceeds to step S02.
Step S02: cutting and numbering the palm print image into blocks to form a palm print image block quality evaluation sequence;
dividing the palm print image into H according to each block sizeblock128 pixels, WblockThe cut is made 128 pixels. Then it totals that
And each image block. There is no overlapping area between each block, and the 324 blocks are numbered in order from left to right, top to bottom.
Step S03: respectively calculating each evaluation index of each palm print image block;
assuming that the currently processed image block is the 150 th image block, calculating evaluation indexes of the palm print image block respectively: (1) the palm print area is in the image block ratio. The index can be obtained by image processing methodblock15564, then calculate formula Cblock15564/(128 × 128) is 0.95. (2) The dryness and wetness of the image block. E of the image block can be obtained firstblock＝176，DRYblock128 |, 176|, 48. (3) Image block definition. Standard deviation S of gray values of all pixels in image blockblock1333. (4) Degree of uniformity of the directions of the lines. Calculating the direction consistency degree of the image block by Gabol filtering method in image processing, CONblock＝202。
Step S04: carrying out comprehensive quality evaluation on all the palm print image blocks by adopting a rank-sum ratio method;
when the four indexes of all 324 palm print image blocks are calculated, step S04 may be started, and the step is further divided into the following sub-steps:
step S041: ranking each evaluation index of the palm print image block;
since the four indexes are used as evaluation bases for quality comprehensive evaluation in this example, the four indexes are ranked for all 324 image blocks, wherein three indexes (1), (3) and (4) are high-priority indexes, and (2) is low-priority indexes. Rank results are given below, where the numbers in parentheses are rank results.
| Block numbering | Cblock | DRYblock | Sblock | CONblock |
| 1 | 0.21(272) | 80(4) | 160(241) | 140(32) |
| 2 | 0.37(180) | 82(6) | 303(181) | 210(23) |
| 3 | 0.33(202) | 120(30) | 411(156) | 71(245) |
| 4 | 0.59(165) | 170(88) | 632(121) | 41(291) |
| 5 | 0.41(172) | 75(3) | 523(133) | 69(249) |
| … | … | … | … | … |
| 323 | 0.66(143) | 70(1) | 872(76) | 82(182) |
| 324 | 0.72(123) | 73(2) | 753(82) | 119(158) |
Step S042: defining a rank and ratio matrix of all palm print image blocks;
step S043: and sequentially calculating the rank and ratio of each palm print image block.
According to the formula
Calculating the rank sum ratio of the palm print image blocks:
Step S05: presetting a plurality of rank and ratio thresholds, counting the number of the palm print image blocks larger than different thresholds, and determining the quality level of each palm print image;
preset rank sum ratio threshold Y10.560 and Y2Sequentially counting the RSR of 324 palm print image blocks as 0.303iValue of Y or more1Number z of (2)135, greater than Y2Number z of (2)2212. Presetting a threshold value Z for comprehensively evaluating the palm print image quality as' excellent1When the palm print image quality is evaluated to be "good", the threshold value is Z2190. The following judgment is made: due to z1<Z1And z is2≥Z2And finally, determining that the result of the comprehensive evaluation of the quality of the current palm print image is good.
While the preferred embodiments of the present invention have been described in detail, the present invention is not limited to the above embodiments, and various changes can be made without departing from the spirit of the present invention within the knowledge of those skilled in the art.
Claims (10)
1. A palm print image quality comprehensive evaluation method based on a rank-sum ratio method is characterized in that: the method comprises the following steps:
step S01: primarily evaluating the quality of the palm print image, and determining that the palm print image which cannot meet the primary evaluation requirement is unqualified in comprehensive evaluation of the quality of the palm print image; wherein the preliminary evaluation comprises two evaluation indexes of whether the palm print image is complete and the size of the palm blank area; the condition that the primary evaluation requirement is not met means that aiming at the condition that any one of the two primary evaluation indexes does not exceed a preset threshold value, the comprehensive evaluation of the palm print image quality is considered to be unqualified, and the subsequent steps are not carried out;
step S02: cutting and numbering the palm print image into blocks to form a palm print image block quality evaluation sequence; the palm print image dicing means that the palm print image is cut into a plurality of image blocks according to a preset size, and the subsequent S03 step calculation is conveniently carried out in a parallel mode;
step S03: respectively calculating each evaluation index of each palm print image block; the evaluation indexes include: (1) the palm print area occupies the image block; (2) the dryness and the humidity of the image block; (3) the image block definition degree; (4) degree of consistency of the palm print line direction;
step S04: carrying out comprehensive quality evaluation on all the palm print image blocks by adopting a rank-sum ratio method; the comprehensive quality evaluation of all the palm print image blocks by adopting the rank-sum ratio method is further divided into the following steps:
step S041: ranking each evaluation index of the palm print image block;
step S042: defining a rank and ratio matrix of all palm print image blocks;
step S043: sequentially calculating the rank and ratio of each palmprint image block;
ranking the evaluation indexes in the step S041, namely, taking the ratio of the palm print area in (1) of the palm print image blocks in the image block; (2) the dryness and the humidity of the image block; (3) the image block definition degree; (4) the consistency degree of the grain line direction, and rank arrangement is carried out on four indexes, wherein three indexes (1), (3) and (4) are high priority and the other index (2) is low priority;
step S05: presetting a plurality of rank and ratio thresholds, counting the number of the palm print image blocks larger than different thresholds, and determining the quality level of each palm print image; presetting a plurality of rank sum ratio thresholds as a set rank sum ratio threshold Y1And Y2Wherein Y is1Value of greater than Y2A value of (d); when the number of the palm print image blocks larger than different thresholds is counted, sequentially and respectively counting the RSR of each palm print image blockiA value of Y or more1And Y2Are respectively z1And z2Wherein is greater than or equal to Y1The ratio of the rank to the RSR of any palm print image blockiThe value is also simultaneously greater than Y2；
The quality grade of each palm print image is determined to be that the threshold value of comprehensive evaluation of palm print image quality as 'excellent' is set as Z1The threshold value for comprehensively evaluating the palm print image quality as "good" is Z2The judgment rule is as follows: if z is1≥Z1Finally, the comprehensive evaluation result of the palm print image quality is determined to be excellent; if z is1＜Z1And z is2≥Z2Finally, the comprehensive evaluation result of the palm print image quality is determined to be good; and finally determining that the palm print image quality comprehensive evaluation result is qualified if the two judgment conditions are not met.
2. The palm print image quality comprehensive evaluation method based on the rank-sum ratio method according to claim 1, characterized in that: in step S01, whether the palm print image is complete is determined as follows: assuming that the height of the palm print image is H pixels and the width is W pixels, whether the palm print image is complete is evaluated according to the ratio of the palm print area to the whole image: cpalm＝Ipalm/(H.times.W), wherein IpalmThe number of pixels in the palm print area in the image.
3. The palm print image quality comprehensive evaluation method based on the rank-sum ratio method according to claim 1, characterized in that: in step S01, the size of the palm space area is determined as follows: considering convenient design of threshold, the reciprocal of the number of pixels in the palm white area is used as an evaluation index, and the calculation formula is Bpalm＝1/Iblank，IblankThe number of the pixels in the palm blank area is the number of the pixels in the palm blank area, wherein the palm blank area refers to an area where the palm center is not collected by the palm print image collecting device, and the reason for the formation is that the palm center is slightly sunken towards the back of the hand compared with the peripheral outline of the palm on the physiological structure, so that the palm center part is not collected by the palm print collecting device.
4. The palm print image quality comprehensive evaluation method based on the rank-sum ratio method according to claim 1, characterized in that: in the step S02: the process of cutting and numbering the palm print image blocks to form a palm print image block quality evaluation sequence is as follows:
assuming the height of each palm print image block as HblockPixel of width WblockThe number of image blocks which can be cut out of the palm print image is
Wherein H and W can be independently selected from HblockAnd WblockAnd (4) dividing, wherein the height of the palm print image is H pixels, and the width of the palm print image is W pixels.
5. The palm print image quality comprehensive evaluation method based on the rank-sum ratio method according to claim 1, characterized in that: in step S03, the method for calculating the image block ratio of the palm print area includes: cblock＝Iblock/(Hblock×Wblock)，IblockThe number of pixels in the palm print area in the image block is as follows, the height of each palm print image block is HblockPixel of width WblockA pixel.
6. The palm print image quality comprehensive evaluation method based on the rank-sum ratio method according to claim 1, characterized in that: in step S03, the dryness and wetness degree of the image block is defined as DRYblock＝|128-Eblock|，EblockIs the desired value, DRY, of the gray values of all pixels in the image blockblockThe higher the palm size, the more dry or wet the palm was during collection.
7. The palm print image quality comprehensive evaluation method based on the rank-sum ratio method according to claim 1, characterized in that: in step S03, the image block definition may be determined by the standard deviation S of the gray-level values of all pixels in the palm print image blockblockEvaluation was carried out.
8. The palm print image quality comprehensive evaluation method based on the rank-sum ratio method according to claim 1, characterized in that: in step S03, the degree of the consistency of the palm print line direction is calculated by filtering the image block using a Gabol filtering method in image processing, so as to obtain the degree of the consistency of the texture direction.
9. The palm print image quality comprehensive evaluation method based on the rank-sum ratio method according to claim 1, characterized in that: in step S042, the process of defining the rank and ratio matrix of all the palm print image blocks is as follows: defining a rank and ratio matrix as
A total of N block images participate in the quality comprehensive evaluation, 4 represents 4 evaluation indexes in the step S041, and any element in the matrix is RijAnd representing the rank ordering result of the j index of the ith palm print image block.
10. The palm print image quality comprehensive evaluation method based on the rank-sum ratio method according to claim 1, characterized in that: in step S043, the process of sequentially calculating the rank and ratio of each palm print image block is as follows: the rank sum ratio for any palm print image block i is:
n denotes the number of image blocks.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710379075.9A CN107169466B (en) | 2017-05-25 | 2017-05-25 | Palm print image quality comprehensive evaluation method based on rank-sum ratio method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710379075.9A CN107169466B (en) | 2017-05-25 | 2017-05-25 | Palm print image quality comprehensive evaluation method based on rank-sum ratio method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107169466A CN107169466A (en) | 2017-09-15 |
| CN107169466B true CN107169466B (en) | 2020-03-31 |
Family
ID=59821631
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710379075.9A Active CN107169466B (en) | 2017-05-25 | 2017-05-25 | Palm print image quality comprehensive evaluation method based on rank-sum ratio method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107169466B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109003259B (en) * | 2018-06-25 | 2022-02-18 | 张威 | Fingerprint quality evaluation method based on line quality expert visual cognition machine learning |
| CN109191457B (en) * | 2018-09-21 | 2022-07-01 | 中国人民解放军总医院 | Pathological image quality validity identification method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102567993A (en) * | 2011-12-15 | 2012-07-11 | 中国科学院自动化研究所 | Fingerprint image quality evaluation method based on main component analysis |
| CN106096513A (en) * | 2016-06-01 | 2016-11-09 | 深圳信炜科技有限公司 | Fingerprint identification method, fingerprint recognition system and electronic equipment |
| CN106295475A (en) * | 2015-05-29 | 2017-01-04 | 北京东方金指科技有限公司 | A kind of based on sum of ranks than the registered fingerprint replacement method of method |
| CN106408552A (en) * | 2015-07-27 | 2017-02-15 | 辽宁省公安厅 | Multi-finger position fingerprint image quality comprehensive evaluation method based on rank sum ratio method |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1238809C (en) * | 2002-09-04 | 2006-01-25 | 长春鸿达光电子与生物统计识别技术有限公司 | Fingerprint identification method as well as fingerprint controlling method and system |
| US7711158B2 (en) * | 2004-12-04 | 2010-05-04 | Electronics And Telecommunications Research Institute | Method and apparatus for classifying fingerprint image quality, and fingerprint image recognition system using the same |
-
2017
- 2017-05-25 CN CN201710379075.9A patent/CN107169466B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102567993A (en) * | 2011-12-15 | 2012-07-11 | 中国科学院自动化研究所 | Fingerprint image quality evaluation method based on main component analysis |
| CN106295475A (en) * | 2015-05-29 | 2017-01-04 | 北京东方金指科技有限公司 | A kind of based on sum of ranks than the registered fingerprint replacement method of method |
| CN106408552A (en) * | 2015-07-27 | 2017-02-15 | 辽宁省公安厅 | Multi-finger position fingerprint image quality comprehensive evaluation method based on rank sum ratio method |
| CN106096513A (en) * | 2016-06-01 | 2016-11-09 | 深圳信炜科技有限公司 | Fingerprint identification method, fingerprint recognition system and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107169466A (en) | 2017-09-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106203395B (en) | Face attribute recognition method based on multitask deep learning | |
| JP2022137166A (en) | System and method of classification of biological particles | |
| JP4420821B2 (en) | Histological assessment | |
| DE60210348T2 (en) | Image identification system | |
| WO2021139258A1 (en) | Image recognition based cell recognition and counting method and apparatus, and computer device | |
| Chang et al. | Gold-standard and improved framework for sperm head segmentation | |
| CN110827306B (en) | Purple soil image segmentation and extraction method based on self-adaptive density peak value clustering | |
| CN116664559B (en) | Machine vision-based memory bank damage rapid detection method | |
| CN105320970B (en) | A kind of potato disease diagnostic device, diagnostic system and diagnostic method | |
| CN101526994A (en) | Fingerprint image segmentation method irrelevant to collecting device | |
| CN107169466B (en) | Palm print image quality comprehensive evaluation method based on rank-sum ratio method | |
| CN110874835B (en) | Crop leaf disease resistance identification method and system, electronic equipment and storage medium | |
| CN114581376A (en) | Automatic sorting method and system for textile silkworm cocoons based on image recognition | |
| CN114627116B (en) | Fabric defect identification method and system based on artificial intelligence | |
| CN101853507B (en) | Cell sorting method for affine propagation clustering | |
| CN113096080B (en) | Image analysis method and system | |
| CN106326914B (en) | A kind of more classification methods of pearl based on SVM | |
| TWI483216B (en) | Analytic system of wafer bin map and analytic method thereof | |
| CN105069403B (en) | A kind of three-dimensional human ear identification based on block statistics feature and the classification of dictionary learning rarefaction representation | |
| Kiratiratanapruk et al. | Silkworm eggs detection and classification using image analysis | |
| CN113096079B (en) | Image analysis system and construction method thereof | |
| CN116596879A (en) | Grape downy mildew self-adaptive identification method based on quantiles of boundary samples | |
| CN114913152A (en) | Machine vision-based wood cutting surface deburring quality evaluation method and system | |
| CN114332027A (en) | Cell uniformity obtaining method and system based on image analysis | |
| WO2020151529A1 (en) | Clothing image classification and image classification methods and apparatuses, and 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 |