CN100375107C - Fingerprint image fragment splicing method - Google Patents
Fingerprint image fragment splicing method Download PDFInfo
- Publication number
- CN100375107C CN100375107C CNB200610001619XA CN200610001619A CN100375107C CN 100375107 C CN100375107 C CN 100375107C CN B200610001619X A CNB200610001619X A CN B200610001619XA CN 200610001619 A CN200610001619 A CN 200610001619A CN 100375107 C CN100375107 C CN 100375107C
- Authority
- CN
- China
- Prior art keywords
- image
- fingerprint
- fingerprint image
- similarity
- joining method
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 37
- 239000012634 fragment Substances 0.000 title description 13
- 238000012545 processing Methods 0.000 claims abstract description 15
- 230000008569 process Effects 0.000 claims abstract description 12
- 230000009467 reduction Effects 0.000 claims abstract description 9
- 238000004364 calculation method Methods 0.000 claims abstract description 5
- 238000005304 joining Methods 0.000 claims description 12
- 238000012217 deletion Methods 0.000 claims description 2
- 230000037430 deletion Effects 0.000 claims description 2
- 238000012795 verification Methods 0.000 claims description 2
- 238000000605 extraction Methods 0.000 abstract description 7
- 238000001914 filtration Methods 0.000 abstract description 3
- 230000003247 decreasing effect Effects 0.000 abstract 1
- 238000003860 storage Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 3
- 239000004744 fabric Substances 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 238000000151 deposition Methods 0.000 description 2
- 238000013519 translation Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000003825 pressing Methods 0.000 description 1
- 238000006748 scratching Methods 0.000 description 1
- 230000002393 scratching effect Effects 0.000 description 1
Images
Landscapes
- Collating Specific Patterns (AREA)
Abstract
The present invention relates to a method controlling the processing process of fingerprint images, particularly to a fingerprint image section splicing method. Compared with the prior art, the present invention has the advantages that the purpose of reducing the storing space of the spliced fingerprint images is obtained by reducing the fingerprint image sections to be spliced, and then the fingerprint image sections after reduction are spliced. In addition, good fingerprint identifying algorithms greatly and finely process original images, such as direction calculation, filtration, characteristic extraction, false characteristic judgment, match, etc. The fingerprint images are reduced, and the calculating amount of the image processing and the characteristic extraction is simultaneously decreased. Thus, for the algorithm which realizes good fingerprint identification on the hardware with a limited calculating ability, program optimization and calculating amount reduction are important under the condition that characteristic points are not lost.
Description
Technical field
The present invention relates to a kind of fingerprint image processing procedure control method, is a kind of joining method of fingerprint image fragment specifically.
Technical background
Along with the continuous development of biological identification technology and perfect, constantly be developed about the product of fingerprint recognition identity.Nearly 2 years, the fingerprint collecting chip of fingerprint image appearred forming by scan mode.Because area is little, so the cost of this class fingerprint collecting chip also is significantly less than touch acquisition chip and optically detecting instrument.Sound attitude power consumption is also all very low, is used widely recently.This class Acquisition Instrument in acquisition principle is: gather the image of a frame one frame, by image mosaic some frame fingerprint image fragments are spliced into a complete fingerprint image then.General view picture fingerprint image all has dozens or even hundreds of kilobyte, and the storage space that not only takies is also very big, and calculated amount is very huge simultaneously.And all be that original image has been carried out a lot of careful processing for some good algorithm for recognizing fingerprint, such as asking direction, filtering, pseudo-characteristic is judged in feature extraction, coupling or the like.Thereby, on the relatively limited hardware of computing power, realizing a good algorithm for recognizing fingerprint, optimizer, reduce calculated amount and just seem particularly important.
The little function of useful several calculated amount of method that generally reduces calculated amount fits the function of calculation of complex, replaces floating-point operation with fixed point calculation, or in advance all values is all calculated, and makes a table, and table look-at is obtained a result in the time of will calculating then.For preceding two kinds of methods, portion reduces computational accuracy, and the result is had certain influence, and the 3rd kind then is to have sacrificed a large amount of storage spaces.These algorithms all do not reduce storage space.
Summary of the invention
The present invention has overcome above-mentioned shortcoming, provides a kind of and reduce picture size in the process of fingerprint image splicing, and then reduce the joining method of the fingerprint image segment of image storage space and computational processing.
The present invention solves the technical scheme that its technical matters takes: a kind of joining method of fingerprint image segment, fingerprint image segment to be spliced is reduced processing, and the fingerprint image segment after will reducing is again spliced, and comprises the steps:
1) position, overlapping region of described current fingerprint image segment to be spliced of calculating and previous frame image segments;
2) determine the position of current stitching image segment in entire image;
3) image segments is reduced processing;
4) image mosaic after will reducing becomes general image.
Described step 1) can realize by following process:
A) in first frame and second frame, choose two image blocks respectively, calculate the highest zone of similarity, judge the coincidence zone of trip from smallest blocks to the identical size between the largest block.
B) to overlapping the highest zone of similarity in the trim process that the zone is listed as, the verification that is listed as.
The pixel that can be odd-numbered line and odd column in the described image segments of deletion is handled in described reduction, or deletes the pixel of even number line and even column in the described image segments.
Can comprise at least one image line in described each image block, the similarity of described image block is the mean value of the similarity of each image line in each image block.
The calculation of similarity degree of described image line can adopt the distance function to each pixel gray-scale value.
The distance function of described gray-scale value can comprise the variance calculating of gray-scale value.
In the described step 4), in the image segments after the described reduction, in the zone of the previous frame doubling of the image, splice processing again behind the mean value of the overlapping desirable gray-scale value of pixel in each pixel and the previous frame image.
Compared with prior art the invention has the beneficial effects as follows: by fingerprint image fragment to be spliced is reduced processing, fingerprint image fragment after will reducing again splices, thereby the purpose of the storage space of fingerprint image after realizing reducing to splice, and, all be that original image has been carried out a lot of careful processing for good algorithm for recognizing fingerprint, such as asking direction, filtering, pseudo-characteristic is judged in feature extraction, coupling or the like.When having reduced fingerprint image, also reduced the calculated amount in Flame Image Process and the characteristic extraction procedure, thereby, for on the relatively limited hardware of computing power, realizing a good algorithm for recognizing fingerprint, under the prerequisite of not losing unique point, optimizer, reduce calculated amount and just seem particularly important.
Description of drawings
Fig. 1 is a workflow diagram of the present invention;
Fig. 2 is an old frame fingerprint image fragment;
Fig. 3 is a new frame fingerprint image fragment;
Fig. 4 is for adding up the overlapping region synoptic diagram by row;
Fig. 5 laterally finely tunes synoptic diagram for the overlapping region;
Fig. 6 is the original image fragment of normal splicing;
Fig. 7 is for pressing the spliced downscaled images fragment of the present invention.
Embodiment
When carrying out fingerprint collecting by sweep type fingerprint sensor, finger is scratching on described sensor, described sensor collects the multiframe fingerprint image successively, Fig. 2, Fig. 3 are respectively the adjacent two frame fingerprint image fragments that priority wherein collects, by setting short acquisition time at interval, making has an identical or very approximate block that is considered as coincideing in the adjacent two two field picture fragments, this identical degree is weighed by similarity, and the foundation of measurement is that the gray-scale value of partial-pixel is arranged is identical or close to two continuous two field pictures.Described similarity is the distance function of corresponding pixel points gray-scale value, and the described distance function that adopts in the present embodiment is the variance computing to two corresponding pixel points gray-scale values, as:
The similarity of definition image line: adopt row corresponding, each gets delegation in two two field pictures exactly, is used as this two similarity of going by the quadratic sum of the difference of the gray-scale value of each corresponding picture element.Therefore described, more little corresponding to the value of similarity, similarity is high more, just illustrates that two row correspondence image are just identical more.
The similarity of definition image block: it is capable of an image block to be exactly that two two field pictures are respectively got continuous n, calculates the similarity of corresponding row in two image blocks respectively, and then similarity value of each row being added up to average is used as the similarity of these two blocks of images.Be not difficult to find out that capable similarity is the special circumstances when n is 1.
In the process of image segments splicing, two field picture is carried out suitable reduction, not only reduced the storage space of original image, and further reduced calculated amount.Below in conjunction with as among Fig. 1, specifically describe image mosaic process in the present embodiment:
At first, the size of supposing every frame is 8 row, 280 row, and normal spliced image size is for should be 320 * 400, and this is the size of original image.First to file two block sizes are 280 * 8 memory headroom, be used for depositing two adjacent two field pictures, be respectively NewSlice[280 row * 8 row] and OldSlice[280 row * 8 row], apply for space ImageBuf[160 row * 200 row of one 400/2 row, 320/2 row again] be used for depositing the image that splices well and dwindled.
At first, gather first two field picture, be put among the NewSlice, as step 101, owing to be the first initial frame, need not splice, so directly give up the pixel that is positioned at odd-numbered line and odd column, keep the pixel that is positioned at even number line and even column, because there are 160 points in the delegation among the ImageBuf, and two field picture deletes that the every row in back has 140 points, then allow and dose 10 gray-scale values before and after the every row of two field picture the pixel that is background colour, be 160 points also so just, copied then in preceding 4 row of the correspondence among the ImageBuf, as step 102.As step 103 two field picture among the NewSlice is copied among the OldSlice then, then gather the new frame of a width of cloth, be put among the NewSlice.
Then, newer frame NewSlice and this two frame of old frame OldSlice, the point in new frame in the pairing new frame of mid point of old frame the 7th row of searching.Specific practice is: obtain the quadratic sum of difference of the gray-scale value of the picture element that the gray-scale value of the picture element in the overlapping region overlaps with it earlier, then divided by the number of picture element as the similarity that overlaps the zone.According to the method described above, be not difficult to find out that if two width of cloth images can splice, the similarity that overlaps the zone is necessarily minimum.Require this minimum value, available traversal or dichotomy wait minimizes.
Here employing is roughly judged coincidence zone, the fine setting that is being listed as then by row earlier.As among Fig. 1 104～110 for adopting the zone of new frame that ergodic algorithm tries to achieve and the similarity value minimum of the row of old frame, in step 104, define several variablees earlier, min is minimum similarity value, X is new frame and the overlapping line number of old frame, and Tempvalue is a process variable, and the 7th row of image segments among Fig. 1 is overlapping with X (X=0～6) row of Fig. 2 respectively, as Fig. 4, the line number of overlapping region is respectively 1,2, and 3,4,5,6,7.Obtain the similarity in this coincidence zone of 7, obtain that of minimum.And then as step 111～117 with several pixels of each translation about Fig. 1, wherein as shown in Figure 5, try to achieve this overlapping similarity several times again, the train value Y when getting minimum similarity value is as the judgement that overlaps the zone.Like this, just found the binding site of new frame and old frame, also found the coordinate of the identical back translation of new frame and old frame, as step 118.
Because old frame carried out subtracting and cut and copy to ImageBuf and suffered this moment,,, just abandons the point on the correspondence if being the point of odd-numbered line or odd column so only need coordinate with the above original image of the correspondence of the picture element in the new frame.In determining ImageBuf, also have under the situation of remaining space, as 119, the average gray value that the picture element of new frame and old frame lap is then got both is used as new gray-scale value and copies among the ImageBuf, as step 121, non-overlapping portions just directly copies among the ImageBuf in the new frame, as step 122, the point that is not capped among the ImageBuf of space is still filled with background colour and is covered, as step 123, before duplicating, to note judging the pixel whether excess space ImageBuf boundary is arranged in every image line, if have, can copy to the other end of this row or directly give up, next return in the step 103, the next frame image is carried out above-mentioned same processing, should be spliced at last as the image among Fig. 6, shear, be spliced into as the image among Fig. 7.
The present invention has not only reduced the storage space of original image by the method for above-mentioned reduction two field picture, and has further reduced calculated amount.As everyone knows, when the length and the wide half that all dwindles of image, it is original 1/4 that area just is reduced into, as Fig. 6,7.Usually handle a width of cloth fingerprint image, need several all in addition each picture element of tens times traversal entire image, when image area is reduced into original 1/4 the time, the calculated amount that is reduced in entire image processing and the characteristic extraction procedure is also considerable, because the ratio of precision of Acquisition Instrument is higher, adopts this method not lose unique point.Matching result influence to latter feature extraction and algorithm is little.For realizing the purpose of above-mentioned downscaled images area, be not limited to the pixel of deleting odd-numbered line and odd column as be shown in the examples, perhaps delete it is that the pixel of even number line and even column also is as a same reason.
According to practical situations, also can reduce each image segments earlier, carry out the splicing of fingerprint image by the overlapping region and the position in general image of calculating between the image segments of reduction back again, adopt this embodiment to compare, can further reduce calculated amount with a last embodiment.
The gray scale of described each pixel can be 0～255 gradual change value, or 0,1 binary value.The present invention is applicable to also that except that fingerprint collecting other carry out the physical characteristics collecting imaging process with scan mode.
More than the joining method of fingerprint image fragment provided by the present invention is described in detail, used specific case herein principle of the present invention and embodiment are set forth, the explanation of above embodiment just is used for helping to understand method of the present invention and core concept thereof; Simultaneously, for one of ordinary skill in the art, according to thought of the present invention, the part that all can change in specific embodiments and applications, in sum, this description should not be construed as limitation of the present invention.
Claims (7)
1. the joining method of a fingerprint image segment, it is characterized in that: fingerprint image segment to be spliced is reduced processing, and the fingerprint image segment after will reducing is again spliced, and comprises the steps:
1) position, overlapping region of described current fingerprint image segment to be spliced of calculating and previous frame image segments;
2) determine the position of current stitching image segment in entire image;
3) image segments is reduced processing;
4) image mosaic after will reducing becomes general image.
2. the joining method of fingerprint image segment according to claim 1 is characterized in that: described step 1) realizes by following process:
A) in first frame and second frame, choose two image blocks respectively, calculate the highest zone of similarity, judge the coincidence zone of trip from smallest blocks to the identical size between the largest block.
B) to overlapping the highest zone of similarity in the trim process that the zone is listed as, the verification that is listed as.
3. the joining method of fingerprint image segment according to claim 1 and 2 is characterized in that: described reduction is treated to the pixel of odd-numbered line and odd column in the described image segments of deletion, or deletes the pixel of even number line and even column in the described image segments.
4. the joining method of fingerprint image segment according to claim 2 is characterized in that: comprise at least one image line in described each image block, the similarity of described image block is the mean value of the similarity of each image line in each image block.
5. the joining method of fingerprint image segment according to claim 4 is characterized in that: the calculation of similarity degree of described image line adopts the distance function to each pixel gray-scale value.
6. the joining method of fingerprint image segment according to claim 5 is characterized in that: the distance function of described gray-scale value comprises the variance calculating of gray-scale value.
7. the joining method of fingerprint image segment according to claim 1 and 2, it is characterized in that: in the described step 4), in the image segments after the described reduction, in the zone of the previous frame doubling of the image, overlapping pixel splices processing after getting the mean value of gray-scale value again in each pixel and the previous frame image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB200610001619XA CN100375107C (en) | 2006-01-18 | 2006-01-18 | Fingerprint image fragment splicing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB200610001619XA CN100375107C (en) | 2006-01-18 | 2006-01-18 | Fingerprint image fragment splicing method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1804862A CN1804862A (en) | 2006-07-19 |
| CN100375107C true CN100375107C (en) | 2008-03-12 |
Family
ID=36866874
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB200610001619XA Expired - Fee Related CN100375107C (en) | 2006-01-18 | 2006-01-18 | Fingerprint image fragment splicing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100375107C (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11163970B1 (en) | 2020-06-16 | 2021-11-02 | Google Llc | Optical fingerprint system with varying integration times across pixels |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103226831B (en) * | 2013-05-02 | 2014-09-24 | 天津大学 | Image Matching Method Using Block Boolean Operation |
| CN104318222B (en) * | 2014-11-14 | 2019-10-11 | 深圳市汇顶科技股份有限公司 | Detection method and device for fingerprint detection |
| CN105447436B (en) * | 2014-12-19 | 2017-08-04 | 比亚迪股份有限公司 | Fingerprint recognition system and fingerprint identification method and electronic equipment |
| CN109409387B (en) * | 2018-11-06 | 2022-03-15 | 深圳增强现实技术有限公司 | Acquisition direction determining method and device of image acquisition equipment and electronic equipment |
| CN112085650A (en) * | 2020-09-09 | 2020-12-15 | 南昌虚拟现实研究院股份有限公司 | Image processing method, image processing device, storage medium and computer equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002335396A (en) * | 2002-02-01 | 2002-11-22 | Sharp Corp | Image compression data processing apparatus and image display method |
| US20030126448A1 (en) * | 2001-07-12 | 2003-07-03 | Russo Anthony P. | Method and system for biometric image assembly from multiple partial biometric frame scans |
| US20040114784A1 (en) * | 2002-11-12 | 2004-06-17 | Fujitsu Limited | Organism characteristic data acquiring apparatus, authentication apparatus, organism characteristic data acquiring method, organism characteristic data acquiring program and computer-readable recording medium on which the program is recorded |
| CN1622632A (en) * | 2004-12-27 | 2005-06-01 | 北京中星微电子有限公司 | A video decoding method for implementing zoom |
| US20060002596A1 (en) * | 2004-07-05 | 2006-01-05 | Nec Infrontia Corporation | Fingerprint reading method, fingerprint reading system and program |
-
2006
- 2006-01-18 CN CNB200610001619XA patent/CN100375107C/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030126448A1 (en) * | 2001-07-12 | 2003-07-03 | Russo Anthony P. | Method and system for biometric image assembly from multiple partial biometric frame scans |
| JP2002335396A (en) * | 2002-02-01 | 2002-11-22 | Sharp Corp | Image compression data processing apparatus and image display method |
| US20040114784A1 (en) * | 2002-11-12 | 2004-06-17 | Fujitsu Limited | Organism characteristic data acquiring apparatus, authentication apparatus, organism characteristic data acquiring method, organism characteristic data acquiring program and computer-readable recording medium on which the program is recorded |
| US20060002596A1 (en) * | 2004-07-05 | 2006-01-05 | Nec Infrontia Corporation | Fingerprint reading method, fingerprint reading system and program |
| CN1622632A (en) * | 2004-12-27 | 2005-06-01 | 北京中星微电子有限公司 | A video decoding method for implementing zoom |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11163970B1 (en) | 2020-06-16 | 2021-11-02 | Google Llc | Optical fingerprint system with varying integration times across pixels |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1804862A (en) | 2006-07-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100375107C (en) | Fingerprint image fragment splicing method | |
| EP1986446B1 (en) | Image processing apparatus and image processing method | |
| US6219462B1 (en) | Method and apparatus for performing global image alignment using any local match measure | |
| CN101646022B (en) | Image splicing method and system thereof | |
| CN102077244A (en) | Method and apparatus for filling in masked domains of a depth map or disparity map estimated based on at least two images | |
| CN1130847A (en) | Apparatus for encoding contour of regions contained in video signal | |
| CN112991469B (en) | A method, device and computer-readable storage medium for face brightness compensation | |
| CN107845066B (en) | Method and device for urban remote sensing image stitching based on piecewise affine transformation model | |
| JP4110501B2 (en) | Random pattern generation apparatus and method, distance image generation apparatus and method, and program providing medium | |
| CN114005120B (en) | License plate character cutting method, license plate recognition method, device, equipment and storage medium | |
| Olsen | Stereo correspondence by surface reconstruction | |
| CN118015190A (en) | Autonomous construction method and device of digital twin model | |
| JP2001194126A (en) | Apparatus and method for measuring three-dimensional shape and program providing medium | |
| Strecha et al. | Motion—Stereo Integration for Depth Estimation | |
| IL123566A (en) | Detecting of relief contours in a pair of stereoscopic images | |
| Stent et al. | Precise deterministic change detection for smooth surfaces | |
| US4783831A (en) | Method for producing a standard pattern for pattern matching | |
| JP3537616B2 (en) | Motion estimation method by computer for temporally consecutive image pixels of video sequence | |
| US6650362B1 (en) | Movement detecting apparatus with feature point extractor based on luminance gradient in current frame | |
| CN118279868A (en) | Method for mutually promoting non-registered infrared visible light image fusion and vehicle target detection | |
| CN117218067A (en) | Complex scene industrial defect identification method combining multi-mode sensing monitoring | |
| CN105677669A (en) | Sorting method and device for panoramic splicing images | |
| CN103714524A (en) | Automatic fusion method for cold-rolled strip steel surface images | |
| KR100551825B1 (en) | Image Alignment Method Using Statistical Correlation | |
| CN1340791A (en) | Method and device for execute linear interpotation of three-dimensional pattern reestablishing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C56 | Change in the name or address of the patentee |
Owner name: FEITIAN TECHNOLOGIES CO., LTD. Free format text: FORMER NAME: BEIJING FEITIAN CHENGXIN TECHNOLOGY CO., LTD. |
|
| CP03 | Change of name, title or address |
Address after: 100085 Beijing city Haidian District Xueqing Road No. 9 Ebizal building B block 17 layer Patentee after: Feitian Technologies Co.,Ltd. Address before: 100083, Haidian District, Xueyuan Road, Beijing No. 40 research, 7A building, 5 floor Patentee before: FEITIAN TECHNOLOGIES Co.,Ltd. |
|
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20080312 |