CN102142092A - Automatic path recognition method during intersections of sperm paths and sperm quality analysis method - Google Patents
Automatic path recognition method during intersections of sperm paths and sperm quality analysis method Download PDFInfo
- Publication number
- CN102142092A CN102142092A CN 201110095652 CN201110095652A CN102142092A CN 102142092 A CN102142092 A CN 102142092A CN 201110095652 CN201110095652 CN 201110095652 CN 201110095652 A CN201110095652 A CN 201110095652A CN 102142092 A CN102142092 A CN 102142092A
- Authority
- CN
- China
- Prior art keywords
- sperm
- path
- sperms
- infall
- paths
- 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.)
- Granted
Links
Images
Landscapes
- Investigating Or Analysing Biological Materials (AREA)
- Image Processing (AREA)
Abstract
The invention relates to an automatic path recognition method in the computer assisted sperm analysis during intersections of sperm paths, which comprises the following the path recognition steps during the intersections of the sperm paths that: when N (N is more than or equal to 2) sperms are intersected, n minus 1 cloned sperms which are at the position completely identical with the position of sperms at the intersections are virtualized; the N sperms are correspondingly associated with the sperms at the intersections and the n minus 1 cloned sperms one by one to update the movement speed, the direction and the acceleration of the N sperms entering the intersections; and when the N sperms dissociate from an intersection zone, respective dissociation paths of the N sperms are found out according to the speed and the direction change. The invention adopts a sperm cloning method; the paths of the N sperms are not interrupted in associative sperm generation paths, so that the path identification is more accurate; and when the automatic path recognition method is applied into the human sperm quality analysis, the accuracy rate of sperm recognition and analysis is further improved. When the two sperms are intersected, the correct recognition rate is improved from 0 to 80 percent; and when the three sperms are intersected, the correct recognition rate is improved from 0 to 70 percent.
Description
Technical field
The present invention relates to the outer semen analysis method of computer assisted human body, the automatic identifying method when particularly relating to sperm paths intersect in the area of computer aided semen analysis.
Background technology
Conventional semen analysis is to observe judgement by veteran medical personnel, comprises the microscopic examination of microscopic examination, seminal fluid stereometry and sperm concentration, survival rate.The conventional analysis result can diagnose some serious male sterilities, comprise that all there are identical structure deformity in azoospermatism, necrozoospermia and all sperms, the male sterility that does not for example have the round end sperm of acrosome, but above-mentioned sterility is the small part in the male sterility.Most of infertile patients of Fa Xianing has normal or mildly subnormal conventional semen analysis result clinically.Traditional artificial semen analysis method often has very big subjectivity, because sperm concentration is big, movement velocity is fast, the reviewer is difficult to accomplish accurate analysis at microscopically, and observation very easily causes operator's asthenopia for a long time, laboratory condition, reviewer's level and experience do not wait, the error of testing result is caused in the capital, has also reduced the objectivity and the comparability of report simultaneously.For example, have data to show, only the disparity range of monitoring result between the technician to percentage of motile spermatozoa reaches 20% at least, even 40%.
Because conventional semen analysis subjectivity is strong, accuracy rate is low, poor repeatability, and there is tangible limitation in prediction to fecundity, computer assisted semen analysis (CASA) system requirements arise at the historic moment, can be referring to the record of following document: [1] Xiao Songshan, Fan Shifu, Li Yanfang, Cao Xingwu, Cao Yuai, the research of area of computer aided sperm motility analytic system, JOU RNAL O F T IANJ IN UN IV ERSITY Vol.33 No.6 Nov.2000; [2] Pang Zengshuan, image partition method chooses and the algorithm realization Hebei Academy of Sciences journal, 21 (3), 2004 in the sperm motility energy force detection system.The CASA system is used in sperm quality analytically with the computer technology in modern times and advanced image processing techniques, easy and simple to handle, analyze directly perceived, data are reliable.CASA sperm motility analytic system block diagram as shown in Figure 1, in existing C ASA system, the ultimate principle of sperm Path Recognition is: the sperm sample is placed on the sample stage, utilizes microslide and cover glass to be fixed; Utilize the system controlled by computer image pick-up card, the dynamic micro-image of sperm is captured continuously, the image sequence that obtains is stored in the computing machine, each width of cloth sperm image is carried out pre-service eliminate noise, utilize threshold segmentation method to carry out the sperm Target Recognition, sperm position according to the different images sequence, with distance, travelling speed, traffic direction is reference, obtain the motion path of sperm, and these paths are carried out Screening Treatment, calculate the canonical parameter of sperm motility at last according to motion path, use for the sperm quality analysis.The basic procedure of existing C ASA system sees also shown in Figure 2.
In existing C ASA system during the sperm Path Recognition, utilize distance, travelling speed, traffic direction to carry out sperm association between multiple image more, when having 2 and above sperm paths intersect, in the time period that intersects, a plurality of sperms will overlap, during this period of time image is carried out Threshold Segmentation and can only obtain a sperm target, thereby cause the path of a plurality of sperms in related this step of generation pass of sperm to produce interruption or profiling error, had a strong impact on the raising of recognition correct rate.Path Recognition problem during the sperm paths intersect has become the difficult point problem in the present sperm path analysis.Can be described: [1] Xiao Songshan, Fan Shifu, Li Yanfang, Cao Xingwu, Cao Yuai, the research of area of computer aided sperm motility analytic system, JOU RNAL O F T IANJ IN UN IV ERS ITYVol.33 No.6 Nov.2000 referring to following document; [2] Pang Zengshuan, image partition method chooses and the algorithm realization Hebei Academy of Sciences journal, 21 (3), 2004 in the sperm motility energy force detection system; [3] Yu Jiangyou, Wang Zhiyan, Zhang Yanqing, the design of sperm multi-target detection and track algorithm and realization, computer engineering and design, 2076,31 (9), 2010.
Summary of the invention
The technical problem to be solved in the present invention provides a kind of to the path automatic identifying method during the sperm paths intersect in the CASA system, and Path Recognition accurately when realizing that many sperms intersect improves sperm identification and the accuracy of analyzing.
For solving the problems of the technologies described above, path automatic identifying method when the invention provides sperm paths intersect in a kind of area of computer aided semen analysis, Path Recognition step when comprising the sperm paths intersect is characterized in that, the path identification step is during the sperm paths intersect:
When N sperm intersected, it was identical with the infall sperm but number different N-1 and clone sperm to fictionalize the position; Described N is more than or equal to 2;
The sperm that N enters infall and this sperm of infall and N-1 clone's sperm carry out corresponding one by one related, are individual sperm renewal movement velocity size, direction and the acceleration that enters infall of N;
When N sperm dissociates the zone of intersection,, find out N sperm free path separately according to the variation of its velocity magnitude and direction.
Path Recognition step during the sperm paths intersect comprises:
At T constantly, N sperm reaches the position at sperm Q place jointly;
Traversal number of times to sperm Q is added up, and initial number of times is 0;
When seeking its position in the next frame image one by one, all detect sperm Q, make whenever being detected of sperm Q once just will travel through number of times and add 1 for each sperm of infall;
The sperm that the traversal number of times is surpassed 1 time carries out " clone ", obtains N-1 described clone's sperm.
This sperm of sperm that N enters infall and infall and N-1 clone's sperm carry out corresponding one by one related after, be individual sperm renewal movement velocity size, direction and the acceleration that enters infall of N, computing method are:
The direction of motion of sperm: utilize the coordinate in continuous 3 two field pictures, obtain the horizontal and vertical position distance between frame and the frame, utilize formula
Find the solution angle and ask average;
The movement velocity of sperm: with the coordinate in continuous 3 two field pictures, obtain the horizontal and vertical position distance between frame and the frame, utilize formula
Find the solution distance and ask average, speed unit is a pixel/frame;
The acceleration of motion of sperm: the absolute value of the velocity variations of front and back 2 frames.
Finding out N sperm free route method separately is: according to size, direction and the acceleration of the movement velocity of each sperm of infall, utilize minimum movement speed and direction to change criterion, from the position of sperm Q and clone's sperm, find its position in the next frame image respectively.
Path Recognition step during described sperm paths intersect is:
1) as N sperm P (i) paths intersect, the infall sperm is Q, sets initial value: i=1, the traversal number of times j=0 of sperm Q; Wherein, i is 1 to N;
2) get sperm P (i);
3) judge whether to travel through number of times j 〉=1;
When traveling through number of times is not j 〉=1, then related P (i) and Q; And for j adds 1, j=j+1;
When the traversal number of times is j 〉=1, then clone Q and obtain Q (j), related P (i) and Q (j), and for j adds 1, j=j+1;
4) direction, size, the acceleration of the movement velocity of renewal P (i);
5) whether judge P (i) then also in the zone of intersection,
When P (i) also in the zone of intersection, continue to upgrade direction, size, the acceleration of its movement velocity;
When P (i) not in the zone of intersection, then seek next position, and calculate the free path of P (i) by the variation of velocity magnitude and direction for P (i); The path extraction of P (i) is finished;
6) judge whether i＞N then,
If then the path extraction of all intersection sperms is finished;
If not, then return step 2), execution in step 2 once more) to step 6).
Simultaneously, the present invention also provides a kind of sperm quality analysis method, the path automatic identifying method when comprising sperm paths intersect in the sharp described area of computer aided semen analysis.
Movement velocity and direction and variation thereof when the present invention takes all factors into consideration the sperm intersection, Path Recognition when adopting the method for clone's (virtual) sperm to realize that many sperms intersect, when a plurality of sperms intersect, no longer interrupt in the path of a plurality of sperms in related this step of generation pass of sperm, make Path Recognition more accurate, further improved sperm identification and the accuracy of analyzing.Adopt method of the present invention when 2 sperms intersect, correct recognition rata is increased to when reaching 80%, 3 sperm and intersecting from 0, and correct recognition rata reaches 70%.
Description of drawings
The present invention is described in further detail below in conjunction with accompanying drawing:
Fig. 1 is the sperm motility analytic system block diagram of background technology.
Fig. 2 is the basic flow sheet of the sperm Path Recognition of background technology.
Fig. 3 is the many sperms of the present invention flow charts in the path identification method when intersecting.
Fig. 4 is a path identification method synoptic diagram when two sperm paths intersect: the state when Fig. 4 a represents the sperm paths intersect, Fig. 4 b are illustrated in that infall is virtual has cloned a sperm; Fig. 4 c represents the path status of sperm when infall leaves.Wherein, sperm C overlaps with sperm D.
Embodiment
Path automatic identifying method in the area of computer aided semen analysis of the present invention during the sperm paths intersect is: when two above sperms intersect, intersecting the position that takes place, clone (virtual) source is in a plurality of sperm targets of same position; In this position, it is related one by one that a plurality of sperms that enter simultaneously and the sperm that fictionalizes carry out, and guaranteed the integrality in path separately, can calculate movement velocity size, direction and acceleration that a plurality of sperms enter infall; When the free zone of intersection of sperm, the variation by velocity magnitude and direction draws free path separately, thereby finishes the identification of sperm crossedpath.
The recognizer that a plurality of sperms and 2 sperms intersect under the situation has general character, and the situation with 2 sperm paths intersect describes below.
Path identification method signal when as shown in Figure 4, being two sperm paths intersect enumerating.In Fig. 4 a, there are sperm A and sperm B,
1) at T constantly, A and B reach the position at sperm Q place jointly, shown in Fig. 4 a.Notice that Q is the result of sperm A and sperm B paths intersect.
2) the traversal number of times (promptly using number of times) of sperm C is added up, initial number of times is 0.When seeking its position in the next frame image, detect Q, make the traversal number of times of Q add 1 for the A sperm.When seeking position in the next frame image, will detect Q equally, make that the traversal number of times of Q is 2 for the B sperm.
3) sperm of traversal number of times above 1 time carried out " clone ".Number different sperm D but that is: position of computer virtual and Q are identical, shown in Fig. 4 b.Set up sperm A and Q, B is related with the path of D, size, direction and the acceleration of related update sperm A that utilization is set up and the movement velocity of B.Computing formula is:
The direction of motion of sperm: utilize the coordinate in continuous 3 two field pictures, obtain the horizontal and vertical position distance between frame and the frame, utilize formula
Find the solution angle and ask average.
The movement velocity of sperm: with the coordinate in continuous 3 two field pictures, obtain the horizontal and vertical position distance between frame and the frame, utilize formula
Find the solution distance and ask average, speed unit is a pixel/frame.
The acceleration of motion of sperm: the absolute value of the velocity variations of front and back 2 frames.
4) A and B intersect and begin to leave after continuing for some time, leave when taking place, the size of the movement velocity of sperm A, the B that obtains according to step 3), direction and acceleration, utilize minimum movement speed and direction to change criterion,, find its position in the next frame image respectively from Q and D position, be respectively F and E, shown in Fig. 4 c.
Thereby when having finished two sperm paths intersect of sperm A, B, the Path Recognition of sperm A, B.
When the noise of photographic images is too big, perhaps less the or part of sperm A and B intersecting angle is when binding, and this algorithm identified rate descends.
Concrete steps are described below step in conjunction with this process flow diagram as shown in Figure 3:
1) as N sperm P (i) (i is 1,2 ..., N) paths intersect, the infall sperm is Q, is variable i, j sets initial value: i=1, the traversal number of times j=0 of sperm Q.
2) get sperm P (i);
3) judge whether to travel through number of times j 〉=1;
When traveling through number of times is not j 〉=1, then related P (i) and Q; And for j adds 1, j=j+1;
When the traversal number of times is j 〉=1, then clone (virtual) Q and obtain Q (j), related P (i) and Q (j), and for j adds 1, j=j+1;
4) direction, size, the acceleration of the movement velocity of renewal P (i);
5) whether judge P (i) then also in the zone of intersection,
When P (i) also in the zone of intersection, continue to upgrade direction, size, the acceleration of its movement velocity,
When P (i) not in the zone of intersection (promptly free the zone of intersection), then seek next position, and calculate the free path of P (i) by the variation of velocity magnitude and direction for P (i); The path extraction of P (i) is finished;
6) judge whether i＞N then,
If then the path extraction of all intersection sperms is finished;
If not, then return step 2), execution in step 2 once more)～step 6).Promptly get P (i),,, mean that the path extraction of N sperm of intersection is all finished up to i＞N for next sperm extracts the path.
This method means that when getting sperm P (1), the traversal number of times is 0, related P (1) and Q, at this moment, j=0+1=1; Upgrade direction, size, the acceleration of the movement velocity of P (1),, then seek the next position of P (1), and calculate the free path of P (1) by the variation of velocity magnitude and direction up to the free zone of intersection of P (1); The path extraction of P (1) is finished.Intersect as two or more sperms, second sperm P (2), the traversal number of times is 1, related P (2) and the Q (1) that clones, and j=1+1=2, direction, size, the acceleration of the movement velocity of renewal P (2) are up to the free zone of intersection of P (2), then seek the next position of P (2), and calculate the free path of P (2) by the variation of velocity magnitude and direction; The path extraction of P (2) is finished.The 3rd sperm P (3), the traversal number of times is 2, related P (3) and the Q (2) that clones, and j=2+1=3, upgrade direction, size, the acceleration of the movement velocity of P (3), up to the free zone of intersection of P (3), then seek the next position of P (3), and calculate the free path of P (3) by the variation of velocity magnitude and direction; The path extraction of P (3) is finished .... by that analogy, all finish up to the path extraction of the sperm that intersects.
Path automatic identifying method in the area of computer aided semen analysis of the present invention during the sperm paths intersect can be applicable to during the outer sperm quality of human body analyzes.
Claims (6)
1. the path automatic identifying method during the sperm paths intersect in the area of computer aided semen analysis, the Path Recognition step when comprising the sperm paths intersect is characterized in that, the path identification step is during the sperm paths intersect:
When N sperm intersected, it was identical with the infall sperm but number different N-1 and clone sperm to fictionalize the position; Described N is more than or equal to 2;
The sperm that N enters infall and this sperm of infall and N-1 clone's sperm carry out corresponding one by one related, are individual sperm renewal movement velocity size, direction and the acceleration that enters infall of N;
When N sperm dissociates the zone of intersection,, find out N sperm free path separately according to the variation of its velocity magnitude and direction.
2. the path automatic identifying method during sperm paths intersect according to claim 1 is characterized in that the Path Recognition step during the sperm paths intersect comprises:
At T constantly, N sperm reaches the position at sperm Q place jointly;
Traversal number of times to sperm Q is added up, and initial number of times is 0;
When seeking its position in the next frame image one by one, all detect sperm Q, make whenever being detected of sperm Q once just will travel through number of times and add 1 for each sperm of infall;
The sperm that the traversal number of times is surpassed 1 time carries out " clone ", obtains N-1 described clone's sperm.
3. the path automatic identifying method during sperm paths intersect according to claim 1, it is characterized in that, this sperm of sperm that N enters infall and infall and N-1 clone sperm carry out corresponding one by one related after, for N sperm that enters infall upgrades movement velocity size, direction and acceleration, computing method are:
The direction of motion of sperm: utilize the coordinate in continuous 3 two field pictures, obtain the horizontal and vertical position distance between frame and the frame, utilize formula
Find the solution angle and ask average;
The movement velocity of sperm: with the coordinate in continuous 3 two field pictures, obtain the horizontal and vertical position distance between frame and the frame, utilize formula
Find the solution distance and ask average, speed unit is a pixel/frame;
The acceleration of motion of sperm: the absolute value of the velocity variations of front and back 2 frames.
4. the path automatic identifying method during sperm paths intersect according to claim 1, it is characterized in that, finding out N sperm free route method separately is: according to size, direction and the acceleration of the movement velocity of each sperm of infall, utilize minimum movement speed and direction to change criterion, from the position of sperm Q and clone's sperm, find its position in the next frame image respectively.
5. the path automatic identifying method during according to each described sperm paths intersect of claim 1-4 is characterized in that, the Path Recognition step during described sperm paths intersect is:
1) as N sperm P (i) paths intersect, the infall sperm is Q, sets initial value: i=1, the traversal number of times j=0 of sperm Q; Wherein, i is 1 to N;
2) get sperm P (i);
3) judge whether to travel through number of times j 〉=1;
When traveling through number of times is not j 〉=1, then related P (i) and Q; And for j adds 1, j=j+1;
When the traversal number of times is j 〉=1, then clone Q and obtain Q (j), related P (i) and Q (j), and for j adds 1, j=j+1;
4) direction, size, the acceleration of the movement velocity of renewal P (i);
5) whether judge P (i) then also in the zone of intersection,
When P (i) also in the zone of intersection, continue to upgrade direction, size, the acceleration of its movement velocity;
When P (i) not in the zone of intersection, then seek next position, and calculate the free path of P (i) by the variation of velocity magnitude and direction for P (i); The path extraction of P (i) is finished;
6) judge whether i＞N then,
If then the path extraction of all intersection sperms is finished;
If not, then return step 2), execution in step 2 once more) to step 6).
6. a sperm quality analysis method is characterized in that, comprises the path automatic identifying method when profit requires sperm paths intersect in each described area of computer aided semen analysis of 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110095652 CN102142092B (en) | 2011-04-15 | 2011-04-15 | Automatic path recognition method during intersections of sperm paths and sperm quality analysis method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110095652 CN102142092B (en) | 2011-04-15 | 2011-04-15 | Automatic path recognition method during intersections of sperm paths and sperm quality analysis method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102142092A true CN102142092A (en) | 2011-08-03 |
| CN102142092B CN102142092B (en) | 2013-03-20 |
Family
ID=44409589
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201110095652 Active CN102142092B (en) | 2011-04-15 | 2011-04-15 | Automatic path recognition method during intersections of sperm paths and sperm quality analysis method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102142092B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007017417A (en) * | 2005-07-05 | 2007-01-25 | Yasuo Nakane | Sperm activity evaluator |
| CN101726578A (en) * | 2008-10-27 | 2010-06-09 | 深圳科瑞克医疗器械有限公司 | Microfluidic biological chip sperm quality analyser |
-
2011
- 2011-04-15 CN CN 201110095652 patent/CN102142092B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007017417A (en) * | 2005-07-05 | 2007-01-25 | Yasuo Nakane | Sperm activity evaluator |
| CN101726578A (en) * | 2008-10-27 | 2010-06-09 | 深圳科瑞克医疗器械有限公司 | Microfluidic biological chip sperm quality analyser |
Non-Patent Citations (3)
| Title |
|---|
| 《第九届全国图象图形学学术会议论文集》 19980531 匡宇等 ***运动图象的多目标检测与分割 209-214 , * |
| 《计算机工程与设计》 20100930 于江有等 ***多目标检测和跟踪算法的设计与实现 2076-2079 1-6 , * |
| 《计算机应用》 19970131 王开平等 MIAS计算机辅助***运动分析***的设计和实现 8-10 1-6 第17卷, 第1期 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102142092B (en) | 2013-03-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Hu et al. | Joint monocular 3D vehicle detection and tracking | |
| CN110362210B (en) | Human-computer interaction method and device integrating eye movement tracking and gesture recognition in virtual assembly | |
| EP4016255A1 (en) | Method and apparatus for identifying gaze behavior in three-dimensional space, and storage medium | |
| CN105574543B (en) | A kind of vehicle brand type identifier method and system based on deep learning | |
| CN112750150B (en) | Vehicle flow statistical method based on vehicle detection and multi-target tracking | |
| TWI448977B (en) | Method and apparatus for video analytics based object counting | |
| CN104317391A (en) | Stereoscopic vision-based three-dimensional palm posture recognition interactive method and system | |
| CN108986138A (en) | Method for tracking target and equipment | |
| JP2017535830A (en) | Method and system for temporal segmentation for gesture analysis | |
| EP3035235B1 (en) | Method for setting a tridimensional shape detection classifier and method for tridimensional shape detection using said shape detection classifier | |
| CN104237121B (en) | A kind of motility of sperm method for testing performance | |
| JP2013196684A (en) | Object counting method and object counting device | |
| CN101726498B (en) | Intelligent detector and method of copper strip surface quality on basis of vision bionics | |
| CN103677274A (en) | Interactive projection method and system based on active vision | |
| CN105719311A (en) | Vehicle System For Detecting Object And Operation Method Thereof | |
| CN109658385A (en) | Eye fundus image judgment method and equipment | |
| JP2017077413A5 (en) | ||
| CN105930808A (en) | Moving object tracking method based on vector boosting template updating | |
| CN105765608A (en) | Method and apparatus for eye detection from glints | |
| CN106022249A (en) | Dynamic object identification method, device and system | |
| CN109549619A (en) | Eyeground disk determines method, glaucoma disease diagnostic device and system along width | |
| CN102142092B (en) | Automatic path recognition method during intersections of sperm paths and sperm quality analysis method | |
| CN109615641A (en) | Multiple target pedestrian tracking system and tracking based on KCF algorithm | |
| KR20200062764A (en) | Method and system for segmentation of vessel using deep learning | |
| Tumuklu Ozyer et al. | A hybrid IMM-JPDAF algorithm for tracking multiple sperm targets and motility analysis |
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 |