TWI757861B - Collective tracking and counting device and method - Google Patents
Collective tracking and counting device and method Download PDFInfo
- Publication number
- TWI757861B TWI757861B TW109130580A TW109130580A TWI757861B TW I757861 B TWI757861 B TW I757861B TW 109130580 A TW109130580 A TW 109130580A TW 109130580 A TW109130580 A TW 109130580A TW I757861 B TWI757861 B TW I757861B
- Authority
- TW
- Taiwan
- Prior art keywords
- sample
- moving objects
- collective
- tracking
- collective tracking
- Prior art date
Links
Images
Landscapes
- Investigating Or Analysing Biological Materials (AREA)
- Apparatus Associated With Microorganisms And Enzymes (AREA)
Abstract
Description
本發明係與移動物體有關，尤其是關於一種集體追蹤與計數裝置及方法。The present invention relates to moving objects, and more particularly to a collective tracking and counting device and method.
近年來，低生育率已經成為世界上許多國家所面臨的嚴重問題之一。以台灣為例，2019年台灣的生育率為1.15，在全球200個國家排名中為倒數第二，僅高於南韓。由於生育率往往與男性***品質有關，因此，與生育率有關的***問題亦成為現代社會愈來愈重要的研究項目。In recent years, low fertility has become one of the serious problems faced by many countries in the world. Take Taiwan as an example. In 2019, Taiwan's fertility rate was 1.15, ranking second to last among the 200 countries in the world, only higher than South Korea. Since fertility is often related to the quality of male sperm, sperm problems related to fertility have become an increasingly important research project in modern society.
在醫療院所檢查夫妻的***症問題時，第一項檢驗項目通常是***分析。一般而言，活動***數量(Total Motile Sperm Count, TMSC)與平均曲線速率(Curvilinear Velocity, VCL)通常是判斷***品質的兩項重要特徵參數。其中，活動***數量的大小與自然懷孕率的關聯性高，而平均曲線速率較快的***樣品往往包含較多的健康***。When examining couples for infertility problems at a medical facility, the first test is usually a semen analysis. Generally speaking, Total Motile Sperm Count (TMSC) and average curve velocity (Curvilinear Velocity, VCL) are usually two important characteristic parameters for judging sperm quality. Among them, the size of the motile sperm count was highly correlated with the natural pregnancy rate, and sperm samples with a faster average curve velocity tended to contain more healthy sperm.
醫療院所常會採用電腦輔助***活動性分析儀(CASA, Computer Assisted Sperm Analyzer)來判斷活動***的數量。電腦輔助***活動性分析儀係使用傳統的點追跡技術(Point Tracking Technology)個別追蹤每個***的軌跡，並能夠評估數十種不同的***參數。為了避免由於***樣品的濃度較高導致許多***重疊而容易誤判，世界衛生組織(WHO)建議：在使用電腦輔助***活動性分析儀時，需將***樣品的***濃度稀釋至低於50×106/ml。然而，稀釋***樣品的***濃度不僅會造成使用者的不便，每次稀釋亦會增加量測時的不準確性。Medical institutions often use Computer Assisted Sperm Analyzer (CASA, Computer Assisted Sperm Analyzer) to determine the number of motile sperm. The computer-assisted sperm motility analyzer tracks the trajectory of each sperm individually using traditional Point Tracking Technology and is capable of evaluating dozens of different sperm parameters. In order to avoid misjudgment due to the high concentration of semen samples resulting in the overlapping of many sperm, the World Health Organization (WHO) recommends that when using a computer-assisted sperm motility analyzer, the sperm concentration of the semen sample should be diluted to less than 50×106 /ml. However, diluting the sperm concentration of the semen sample not only causes inconvenience to the user, but also increases the inaccuracy of the measurement with each dilution.
此外，如圖1所示，若以***影片為例，傳統的點追蹤技術包括下列步驟：首先，在輸入***影片的一系列影格序列(步驟S10)後，傳統的點追蹤技術會於單一影格中忽略背景而僅辨認***等有興趣的物體(步驟S12)。接著，傳統的點追蹤技術會分類物體，僅選取單一影格中特定大小的個別***，而忽略影格中其他尺寸太大或太小的雜質(步驟S14)。In addition, as shown in FIG. 1, if a sperm video is taken as an example, the traditional point tracking technology includes the following steps: First, after inputting a series of frame sequences of the sperm video (step S10), the traditional point tracking technology will track a single frame in a single frame. The background is ignored and only interesting objects such as sperm are recognized (step S12). Next, the conventional point tracking technology classifies the objects and selects only individual spermatozoa of a certain size in a single frame, while ignoring other impurities in the frame that are too large or too small (step S14 ).
之後，傳統的點追蹤技術會追跡物體，將兩個影格之間位置座標最接近的***辨認為同一隻***，並紀錄所有個別***在兩個影格之間的移動軌跡座標(步驟S16)。需說明的是，當***樣品的濃度較高導致許多***重疊時，步驟S16需要較多的計算，以區分個別***而不至於混淆。然後，傳統的點追蹤技術會將多個影格之間的所有個別***的座標位置紀錄起來，得到所有個別***移動軌跡的一系列座標點(步驟S18)。最後，傳統的點追蹤技術會根據這些座標點計算出所有個別***的速率，並可得到活動***的總數與平均速率(步驟S19)。Afterwards, the conventional point tracking technology will track the object, identify the sperm with the closest position coordinates between the two frames as the same sperm, and record the movement track coordinates of all the individual sperm between the two frames (step S16 ). It should be noted that when the concentration of the semen sample is high and many sperms overlap, step S16 requires more calculations to distinguish individual sperms without confusion. Then, the conventional point tracking technology records the coordinate positions of all individual sperms between multiple frames, and obtains a series of coordinate points of all individual sperm moving trajectories (step S18 ). Finally, the traditional point tracking technology will calculate the velocity of all individual spermatozoa according to these coordinate points, and obtain the total number and average velocity of motile spermatozoa (step S19).
由上述可知：傳統的點追蹤技術在實際應用中仍存在著花費的時間較長且能分析的物體較少等問題，亟待克服。因此，若能發展出一種簡易的***活動性檢驗裝置，不僅可同時偵測活動***的數量(TMSC)與平均速率(VCL)等重要的特徵參數，而且不需稀釋樣品即可提供***樣品的初步檢驗，方便受試者檢驗其***的品質。It can be seen from the above that the traditional point tracking technology still has problems such as long time consumption and few objects that can be analyzed in practical applications, which need to be overcome urgently. Therefore, if a simple sperm motility test device can be developed, it can not only detect important characteristic parameters such as the number of motile sperm (TMSC) and average velocity (VCL) at the same time, but also provide sperm samples without diluting the samples. Preliminary test, which is convenient for subjects to test the quality of their sperm.
有鑑於此，本發明提出一種集體追蹤與計數裝置及方法，以克服先前技術所遭遇到的問題。In view of this, the present invention proposes a collective tracking and counting device and method to overcome the problems encountered in the prior art.
依據本發明之一具體實施例為一種集體追蹤與計數裝置。於此實施例中，集體追蹤與計數裝置包含光學成像裝置、影像感測器、分析晶片及顯示器。光學成像裝置包含光源，用以發出光線至包含有移動物體的樣品，以產生樣品影像。影像感測器相對於光學成像裝置而設置，用以連續感測不同時間的樣品影像。分析晶片耦接影像感測器，用以對不同時間的樣品影像進行分析後產生分析結果。顯示器耦接分析晶片，用以顯示分析結果。One embodiment according to the present invention is a collective tracking and counting device. In this embodiment, the collective tracking and counting device includes an optical imaging device, an image sensor, an analysis chip, and a display. The optical imaging device includes a light source for emitting light to a sample including a moving object to generate an image of the sample. The image sensor is arranged relative to the optical imaging device, and is used to continuously sense sample images at different times. The analysis chip is coupled to the image sensor for analyzing sample images at different times to generate analysis results. The display is coupled to the analysis chip for displaying analysis results.
於一實施例中，分析晶片係採用差分軌跡法對不同時間的該些樣品影像進行分析以產生分析結果。In one embodiment, the analysis wafer adopts the differential trajectory method to analyze the sample images at different times to generate analysis results.
於一實施例中，分析晶片分析得到樣品中之該些移動物體隨時間累積的差分軌跡面積。In one embodiment, the differential track area accumulated over time for the moving objects in the sample is obtained by analyzing the wafer.
於一實施例中，分析結果包含與樣品中之該些移動物體的活動特性相關的至少一特徵參數。In one embodiment, the analysis result includes at least one characteristic parameter related to the motion characteristics of the moving objects in the sample.
於一實施例中，該至少一特徵參數為活動***數量(Total Motile Sperm Count, TMSC)及/或平均曲線速率(Curvilinear Velocity, VCL)。In one embodiment, the at least one characteristic parameter is Total Motile Sperm Count (TMSC) and/or Curvilinear Velocity (VCL).
依據本發明之另一具體實施例為一種集體追蹤與計數方法。於此實施例中，集體追蹤與計數方法包含下列步驟：(a)提供樣品的視頻序列；(b)採用影格差分技術偵測樣品中的移動物體；(c)隨時間累積得到總差分軌跡面積；以及(d)產生樣品的至少一特徵參數。Another embodiment according to the present invention is a collective tracking and counting method. In this embodiment, the collective tracking and counting method includes the following steps: (a) providing a video sequence of the sample; (b) detecting moving objects in the sample using frame difference technology; (c) accumulating a total difference track over time area; and (d) generating at least one characteristic parameter of the sample.
於一實施例中，該至少一特徵參數係與樣品中之該些移動物體的活動特性相關。In one embodiment, the at least one characteristic parameter is related to motion characteristics of the moving objects in the sample.
於一實施例中，該至少一特徵參數為活動***數量(TMSC)及/或平均曲線速率(VCL)。In one embodiment, the at least one characteristic parameter is motile sperm count (TMSC) and/or mean velocity of curve (VCL).
於一實施例中，樣品的視頻序列包含於不同時間連續感測到的複數個樣品影像，且步驟(c)係採用差分軌跡法對該些樣品影像進行分析。In one embodiment, the video sequence of the sample includes a plurality of sample images continuously sensed at different times, and the step (c) uses the differential trajectory method to analyze the sample images.
於一實施例中，該些樣品影像係由光源發出光線至樣品中之該些移動物體而成像產生。In one embodiment, the sample images are imaged by a light source emitting light to the moving objects in the sample.
相較於先前技術，本發明的集體追蹤與計數裝置及方法係利用影像感測器連續紀錄樣品中之移動物體於不同時間的影像，再透過分析晶片根據差分軌跡法即時呈現樣品中之移動物體的集體移動軌跡，並進一步計算出樣品中之移動物體的數量、平均速率等特徵參數，藉以判定樣品之活動性高低。Compared with the prior art, the collective tracking and counting device and method of the present invention utilize an image sensor to continuously record images of moving objects in a sample at different times, and then present the moving objects in the sample in real time through an analysis chip according to a differential trajectory method. The collective movement trajectory of the sample is further calculated, and the characteristic parameters such as the number and average speed of the moving objects in the sample are further calculated, so as to determine the activity level of the sample.
因此，本發明的集體追蹤與計數裝置及方法不需採用複雜的點追蹤技術並能有效解決樣品中之移動物體彼此重疊的問題，故可應用於高密度樣品之分析，還同時具有步驟簡單、成本低廉、應用範圍廣泛等優點，應具有相當高的市場潛力。Therefore, the collective tracking and counting device and method of the present invention do not need to use complex point tracking technology and can effectively solve the problem of overlapping moving objects in the sample, so it can be applied to the analysis of high-density samples, and has the advantages of simple steps, The advantages of low cost and wide application range should have quite high market potential.
關於本發明之優點與精神可以藉由以下的發明詳述及所附圖式得到進一步的瞭解。The advantages and spirit of the present invention can be further understood from the following detailed description of the invention and the accompanying drawings.
現在將詳細參考本發明的示範性實施例，並在附圖中說明所述示範性實施例的實例。在圖式及實施方式中所使用相同或類似標號的元件/構件是用來代表相同或類似部分。Reference will now be made in detail to the exemplary embodiments of the present invention, examples of which are illustrated in the accompanying drawings. Elements/components using the same or similar numbers in the drawings and the embodiments are intended to represent the same or similar parts.
依據本發明之一具體實施例為一種集體追蹤與計數方法。不同於先前技術中所採用的點追蹤法較為複雜，本發明所提出的集體追蹤與計數方法能夠透過較為簡易的步驟同時分析單一畫面中所包含的3,500個以上的移動物體的特徵參數，例如活動***數量(TMSC)及平均曲線速率(VCL)。A specific embodiment according to the present invention is a collective tracking and counting method. Unlike the point tracking method used in the prior art, which is more complicated, the collective tracking and counting method proposed in the present invention can simultaneously analyze the characteristic parameters of more than 3,500 moving objects contained in a single screen through relatively simple steps, such as activities Sperm count (TMSC) and mean velocity of curve (VCL).
於此實施例中，本發明的集體追蹤與計數方法係採用影格差分(Frame differencing)技術將樣本影像的任兩個時間前後相鄰的影格相減，並將兩影格之間對應像素位置的亮度值相減後取絕對值，並紀錄所有差異絕對值大於一設定閾值的像素位置。這些像素位置即可被定義為移動物體的差分足跡(Differential Footprint, DF)。這些差分足跡可隨著時間累積而成為差分軌跡(Differential Footprint Trajectory, DFT)。至於移動物體的數量與速率則可反映於差分軌跡面積隨時間累積圖上。In this embodiment, the collective tracking and counting method of the present invention adopts the frame differencing technique to subtract any two adjacent frames in time before and after the sample image, and calculates the difference between the corresponding pixel positions between the two frames. After subtracting the luminance values, the absolute values are obtained, and all pixel positions whose absolute values of differences are greater than a preset threshold are recorded. These pixel positions can then be defined as the differential footprint (DF) of the moving object. These differential footprints can be accumulated over time into Differential Footprint Trajectory (DFT). As for the number and velocity of moving objects, it can be reflected on the cumulative graph of the differential track area over time.
請參照圖2A至圖2C，圖2A至圖2C係分別繪示在十倍物鏡下，具有不同活動***數量的***樣本的集體差分軌跡圖。其中，圖2A的活動***數量(TMSC)為130；圖2B的活動***數量(TMSC)為46；圖2C的活動***數量(TMSC)為6。需說明的是，圖2A至圖2C均分別包含複數個***的差分軌跡，且一條線代表單一***的差分軌跡，從深灰色到白色依序代表到0到10秒的***移動軌跡。Please refer to FIGS. 2A to 2C . FIGS. 2A to 2C are respectively collective differential trajectory diagrams of semen samples with different motile sperm counts under a 10x objective lens. Among them, the motile sperm count (TMSC) in Figure 2A was 130; the motile sperm count (TMSC) in Figure 2B was 46; the motile sperm count (TMSC) in Figure 2C was 6. It should be noted that, Figures 2A to 2C respectively include differential trajectories of a plurality of spermatozoa, and a line represents the differential trajectory of a single spermatozoa, from dark gray to white sequentially representing the sperm movement trajectory from 0 to 10 seconds.
接著，請參照圖3，圖3係繪示此實施例中之集體追蹤與計數方法的流程圖。如圖3所示，集體追蹤與計數方法包含下列步驟：Next, please refer to FIG. 3 , which is a flowchart illustrating the collective tracking and counting method in this embodiment. As shown in Figure 3, the collective tracking and counting method consists of the following steps:
步驟S20：提供樣品的視頻序列；Step S20: providing the video sequence of the sample;
步驟S22：採用影格差分技術偵測樣品中的移動物體；Step S22: using the frame difference technology to detect the moving object in the sample;
步驟S24：隨時間累積得到總差分軌跡面積；以及Step S24: Accumulate the total differential track area over time; and
步驟S26：產生樣品的至少一特徵參數。Step S26: Generate at least one characteristic parameter of the sample.
於實際應用中，步驟S22中之「影格差分技術」係將任兩個時間前後相鄰的影格之相對應像素的亮度值相減，再取絕對值，若其大於一預定閾值則判定其為「移動物體」，反之則判定為「非移動物體」。藉此，步驟S22可將所有不動的物體及背景去除，而僅留下「移動物體」(例如活動***)的差分足跡。In practical applications, the "frame difference technique" in step S22 is to subtract the luminance values of the corresponding pixels of any two adjacent frames before and after time, and then take the absolute value. If it is greater than a predetermined threshold, it is determined that it is It is a "moving object", otherwise it is judged as a "non-moving object". In this way, step S22 can remove all stationary objects and backgrounds, and only leave differential footprints of "moving objects" (eg, motile sperm).
此外，步驟S24係透過「差分軌跡法」將連續影格中所有移動***的差分足跡累積成集體移動軌跡(亦即差分軌跡)，並隨時間累積而得到集體移動軌跡面積(亦即總差分軌跡面積)。最後，步驟S26根據集體移動軌跡面積(亦即總差分軌跡面積)隨時間累積值計算出樣品中之所有移動物體的集體資訊(例如活動***數量與平均速率等特徵參數)。In addition, step S24 is to accumulate the differential footprints of all moving sperms in the continuous frame into a collective movement trajectory (ie, differential trajectory) through the "differential trajectory method", and accumulate the collective movement trajectory area (ie, the total differential trajectory area) over time. ). Finally, step S26 calculates the collective information of all moving objects in the sample (such as characteristic parameters such as the number of motile sperm and average velocity) according to the accumulated value of the collective moving track area (ie, the total differential track area) over time.
根據上述可知：本發明的集體追蹤與計數方法所採用的差分軌跡法不需辨認、分類及追蹤個別的移動物體，其欲即時呈現的是樣品中之所有移動物體的集體移動軌跡，而非分別判斷每一個移動物體各自的移動軌跡，因此，即使樣品中之移動物體彼此重疊亦不會造成影響，故本發明的集體追蹤與計數方法不需稀釋樣品，即可快速分析高密度的樣品。From the above, it can be seen that the differential trajectory method adopted by the collective tracking and counting method of the present invention does not need to identify, classify and track individual moving objects, and what it wants to present instantly is the collective moving trajectories of all moving objects in the sample, rather than separate moving objects. The respective moving trajectories of each moving object are judged. Therefore, even if the moving objects in the sample overlap each other, it will not affect each other. Therefore, the collective tracking and counting method of the present invention can quickly analyze high-density samples without diluting the sample.
經比較圖3與圖1後，可明確得知：在提供了樣品的視頻序列之後，先前技術需經過四個步驟S12至S18才能得到樣品的特徵參數，而本發明則僅需經過兩個步驟S22至S24即能得到樣品的特徵參數，故本發明所提出的集體追蹤與計數方法的確能夠有效簡化流程步驟。After comparing FIG. 3 and FIG. 1 , it can be clearly known that after the video sequence of the sample is provided, the prior art needs to go through four steps S12 to S18 to obtain the characteristic parameters of the sample, while the present invention only needs to go through two steps. The characteristic parameters of the sample can be obtained from S22 to S24, so the collective tracking and counting method proposed in the present invention can indeed effectively simplify the process steps.
接著，請參照圖4A至圖4D。圖4A係繪示直徑為d的圓形物體在時間t0至t6的七個連續影格以速率v移動的示意圖；圖4B係繪示圓形物體於任兩相鄰影格之間的差分足跡圖；圖4C係繪示圓形物體的差分軌跡圖；圖4D係繪示圓形物體的差分軌跡面積隨時間累積圖。Next, please refer to FIGS. 4A to 4D . 4A is a schematic diagram showing a circular object with a diameter d moving at a rate v in seven consecutive frames from time t0 to t6; FIG. 4B is a diagram showing the differential footprint of a circular object between any two adjacent frames; FIG. 4C is a graph showing the differential trajectory of the circular object; FIG. 4D is a graph showing the cumulative area of the differential trajectory of the circular object over time.
如圖4A所示，本發明可透過光學成像裝置與影像感測器(見圖6)隨著時間ti(i=0至6)連續紀錄樣品中之直徑為d的移動物體以速率v向右方移動的七個連續影格，且每個影格分別包含複數個像素，每個像素均記錄一亮度值(0至255)。As shown in FIG. 4A , the present invention can continuously record the moving object with the diameter d in the sample through the optical imaging device and the image sensor (see FIG. 6 ) with the time ti (i=0 to 6) to the right at the speed v There are seven consecutive frames that move squarely, and each frame includes a plurality of pixels, and each pixel records a luminance value (0 to 255).
如圖4B所示，可將任兩個時間前後相鄰的影格(例如對應於i=0與i=1的兩影格)之間對應像素位置的亮度值相減後再取絕對值。當兩相鄰影格之間對應的像素位置沒有任何亮度上的改變時，相減為零。這些像素位置上標示為灰階顏色的區域，形成移動物體的差分足跡且其亮度改變量大於設定閾值時，則把該對應的像素位置上用某種灰階顏色標記而形成差分足跡影像DF1至DF6。As shown in FIG. 4B , the luminance values of corresponding pixel positions between any two frames adjacent in time (eg, two frames corresponding to i=0 and i=1) can be subtracted, and then absolute values can be obtained. When there is no change in the brightness of the corresponding pixel positions between two adjacent frames, the subtraction is zero. The areas marked with gray-scale color on these pixel positions form the differential footprint of the moving object and when the brightness change is greater than the set threshold, the corresponding pixel positions are marked with a certain gray-scale color to form differential footprint images DF1 to DF1 to DF6.
上述的該些差分足跡的總和面積愈大，代表樣品中之移動物體愈多或是移動物體的速率愈快。如圖4C所示，接下來可將樣品中之所有移動物體的差分足跡逐一疊加而形成差分軌跡影像DFT1至DFT6。隨著時間的推移，每個差分軌跡都會延伸，使得差分軌跡的總面積會隨著時間而增加。需說明的是，這些差分軌跡可反映樣品中之所有移動物體的移動軌跡，舉例而言，較長的差分軌跡代表物體移動速率較快，而差分軌跡愈多代表移動物體的數量愈多。The larger the total area of the above differential footprints, the more moving objects in the sample or the faster the moving objects. As shown in FIG. 4C , the differential footprints of all moving objects in the sample can be superimposed one by one to form differential trajectory images DFT1 to DFT6 . Over time, each differential trace extends so that the total area of the differential trace increases over time. It should be noted that these differential trajectories can reflect the moving trajectories of all moving objects in the sample. For example, longer differential trajectories represent faster object movement rates, and more differential trajectories represent more moving objects.
如圖4D所示，在差分軌跡面積隨時間累積的曲線上可得到一轉折點，並且根據此轉折點的座標位置即可得到樣品中之所有移動物體的總數量及平均速率，而不需如同先前技術一樣去分別追蹤樣品中之每一個移動物體。As shown in Fig. 4D, a turning point can be obtained on the curve of the accumulation of differential track areas over time, and the total number and average velocity of all moving objects in the sample can be obtained according to the coordinate position of this turning point, without the need for the prior art. Likewise, track each moving object in the sample separately.
接著，如圖5所示，本發明採用差分軌跡法分析樣品所花費的分析時間(約0.6秒)明顯短於傳統的點追蹤法分析樣品所花費的分析時間(3.4秒至6.7秒)，並且本發明採用差分軌跡法能分析1至3,500個移動物體亦明顯多於傳統的點追蹤法僅能分析1至200個移動物體。換言之，相較於傳統的點追蹤法，本發明採用的差分軌跡法花費較少的時間且能分析較多的移動物體。需說明的是，差分軌跡法分析樣品花費的計算時間，與樣品中包含的移動物體數量無關，故能夠不稀釋樣品，快速分析高密度的樣品。Next, as shown in FIG. 5 , the analysis time (about 0.6 seconds) taken by the present invention to analyze the sample by the differential trajectory method is significantly shorter than the analysis time (3.4 seconds to 6.7 seconds) taken by the traditional point tracking method to analyze the sample, and The present invention can analyze 1 to 3,500 moving objects by using the differential trajectory method, which is significantly more than that of the traditional point tracking method, which can only analyze 1 to 200 moving objects. In other words, compared with the traditional point tracking method, the differential trajectory method adopted in the present invention takes less time and can analyze more moving objects. It should be noted that the calculation time spent in analyzing a sample by the differential trajectory method is independent of the number of moving objects contained in the sample, so it can quickly analyze a high-density sample without diluting the sample.
依據本發明之另一具體實施例為一種集體追蹤與計數裝置。不同於先前技術中所採用的點追蹤法較為複雜，本發明所提出的集體追蹤與計數裝置能夠透過較為簡易的步驟同時分析單一畫面中所包含的3,500個以上的移動物體的特徵參數，例如活動***數量(TMSC)及平均曲線速率(VCL)。Another embodiment according to the present invention is a collective tracking and counting device. Different from the complex point tracking method used in the prior art, the collective tracking and counting device proposed in the present invention can simultaneously analyze the characteristic parameters of more than 3,500 moving objects included in a single screen through relatively simple steps, such as activities Sperm count (TMSC) and mean velocity of curve (VCL).
請參照圖6，圖6係繪示此實施例中之集體追蹤與計數裝置的示意圖。如圖6所示，集體追蹤與計數裝置1包含光學成像裝置10、影像感測器12、分析晶片14及顯示器16。光學成像裝置10包含光源100，用以發出光線至包含有移動物體MO的樣品SP，以產生樣品影像。影像感測器12相對於光學成像裝置10而設置，用以連續感測不同時間的樣品影像。分析晶片14耦接影像感測器12，用以對不同時間的樣品影像進行分析後產生分析結果。顯示器16耦接分析晶片14，用以顯示分析結果。Please refer to FIG. 6 , which is a schematic diagram of the collective tracking and counting device in this embodiment. As shown in FIG. 6 , the collective tracking and
於實際應用中，分析晶片14係採用差分軌跡法對不同時間的該些樣品影像進行分析，以得到樣品SP中之該些移動物體MO隨時間累積的差分軌跡面積。分析結果可包含與樣品SP中之該些移動物體MO的活動特性相關的至少一特徵參數，例如活動***數量(TMSC)及/或平均曲線速率(VCL)，但不以此為限。In practical applications, the analysis chip 14 uses the differential trajectory method to analyze the sample images at different times, so as to obtain the differential trajectory areas accumulated over time of the moving objects MO in the sample SP. The analysis result may include at least one characteristic parameter related to the motility characteristics of the moving objects MO in the sample SP, such as, but not limited to, motile sperm count (TMSC) and/or mean velocity of curve (VCL).
相較於先前技術，本發明的集體追蹤與計數裝置及方法係利用影像感測器連續紀錄樣品中之移動物體於不同時間的影像，再透過分析晶片根據差分軌跡法即時呈現樣品中之移動物體的集體移動軌跡，並進一步計算出樣品中之移動物體的數量、平均速率等特徵參數，藉以判定樣品之活動性高低。Compared with the prior art, the collective tracking and counting device and method of the present invention utilize an image sensor to continuously record images of moving objects in a sample at different times, and then present the moving objects in the sample in real time through an analysis chip according to a differential trajectory method. The collective movement trajectory of the sample is further calculated, and the characteristic parameters such as the number and average speed of the moving objects in the sample are further calculated, so as to determine the activity level of the sample.
因此，本發明的集體追蹤與計數裝置及方法不需採用複雜的點追蹤技術並能有效解決樣品中之移動物體彼此重疊的問題，故可應用於高密度樣品之分析，還同時具有步驟簡單、成本低廉、應用範圍廣泛等優點，應具有相當高的市場潛力。Therefore, the collective tracking and counting device and method of the present invention do not need to use complex point tracking technology and can effectively solve the problem of overlapping moving objects in the sample, so it can be applied to the analysis of high-density samples, and has the advantages of simple steps, The advantages of low cost and wide application range should have quite high market potential.
S10~S19:步驟 S20~S26:步驟 ti:時間 d:直徑 v:速率 DF1~DF6:差分足跡影像 bR :第一位置 bF :第二位置 DFT1~DFT6:差分軌跡影像S10~S19: Steps S20~S26: Steps ti: Time d: Diameter v: Velocity DF1~DF6: Differential footprint image b R : First position b F : Second position DFT1~DFT6: Differential track image
本發明所附圖式說明如下： 圖1係繪示傳統的點追蹤物體追蹤與計數方法的流程圖。 圖2A至圖2C係分別繪示具有不同活動***數的***樣本的灰階顏色編碼差分軌跡圖。 圖3係繪示根據本發明之一較佳具體實施例中之集體追蹤與計數方法的流程圖。 圖4A係繪示直徑為d的圓形物體在時間t0至t6的七個連續影格以速率v移動的示意圖。 圖4B係繪示圓形物體於任兩相鄰影格之間的差分足跡圖。 圖4C係繪示圓形物體的差分軌跡圖。 圖4D係繪示圓形物體的差分軌跡面積隨時間累積圖。 圖5係繪示本發明的差分軌跡法與傳統的點追蹤法所花費的分析時間與能夠分析的移動物體數量之比較圖。 圖6係繪示根據本發明之另一較佳具體實施例中之集體追蹤與計數裝置的示意圖。The accompanying drawings of the present invention are described as follows: FIG. 1 is a flowchart illustrating a conventional point tracking object tracking and counting method. 2A to 2C are respectively gray-scale color-coded differential trajectories of semen samples with different motile sperm counts. FIG. 3 is a flowchart illustrating a collective tracking and counting method according to a preferred embodiment of the present invention. FIG. 4A is a schematic diagram showing a circular object with a diameter d moving at a rate v in seven consecutive frames from time t0 to t6. FIG. 4B is a graph showing the differential footprint of a circular object between any two adjacent frames. FIG. 4C is a differential trajectory diagram of a circular object. FIG. 4D is a graph showing the accumulation of differential track areas of a circular object over time. FIG. 5 is a graph showing the comparison between the analysis time spent by the differential trajectory method of the present invention and the conventional point tracking method and the number of moving objects that can be analyzed. FIG. 6 is a schematic diagram illustrating a collective tracking and counting device according to another preferred embodiment of the present invention.
S20~S26:步驟S20~S26: Steps
Claims (9)
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US201962896930P | 2019-09-06 | 2019-09-06 | |
| US62/896,930 | 2019-09-06 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| TW202111725A TW202111725A (en) | 2021-03-16 |
| TWI757861B true TWI757861B (en) | 2022-03-11 |
Family
ID=76035710
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| TW109130580A TWI757861B (en) | 2019-09-06 | 2020-09-07 | Collective tracking and counting device and method |
Country Status (1)
| Country | Link |
|---|---|
| TW (1) | TWI757861B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019125583A1 (en) * | 2017-12-22 | 2019-06-27 | Hillel Llc | Imaging device for measuring sperm motility |
-
2020
- 2020-09-07 TW TW109130580A patent/TWI757861B/en active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019125583A1 (en) * | 2017-12-22 | 2019-06-27 | Hillel Llc | Imaging device for measuring sperm motility |
Non-Patent Citations (1)
| Title |
|---|
| 期刊韕Wei, Sung-Yang, et al. 韕 韕"A collective tracking method for preliminary sperm analysis." Biomedical engineering online 韕18.1 (2019)韕BioMed Central韕2019韕p1-17.韕 * |
Also Published As
| Publication number | Publication date |
|---|---|
| TW202111725A (en) | 2021-03-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Cheng et al. | Decoupled two-stage crowd counting and beyond | |
| CN109145708B (en) | Pedestrian flow statistical method based on RGB and D information fusion | |
| Tek et al. | Computer vision for microscopy diagnosis of malaria | |
| TWI667621B (en) | Face recognition method | |
| US11484752B2 (en) | Jump counting method for jump rope | |
| CN113252568A (en) | Lens surface defect detection method, system, product and terminal based on machine vision | |
| CN106548131A (en) | A kind of workmen's safety helmet real-time detection method based on pedestrian detection | |
| Yao et al. | Automated detection of non-informative frames for colonoscopy through a combination of deep learning and feature extraction | |
| CN108364306B (en) | Visual real-time detection method for high-speed periodic motion | |
| CN101316371A (en) | Flame detecting method and device | |
| TWI757861B (en) | Collective tracking and counting device and method | |
| CN111340841A (en) | Real-time multi-moving-object detection method and system | |
| CN114613006A (en) | Remote gesture recognition method and device | |
| Punitha et al. | Detection of malarial parasite in blood using image processing | |
| TW201404878A (en) | Device for automatically rapidly analyzing biological cells and related method thereof | |
| CN108388883A (en) | A kind of video demographic method based on HOG+SVM | |
| CN111563869B (en) | Stain test method for quality inspection of camera module | |
| CN111582124A (en) | Face recognition and temperature measurement method of thermal infrared imager | |
| Mangrulkar | Retinal image classification technique for diabetes identification | |
| US11681371B2 (en) | Eye tracking system | |
| Schiele et al. | Comparison of deep learning-based image segmentation methods for the detection of voids in X-ray images of microelectronic components | |
| CN106446837B (en) | A kind of detection method of waving based on motion history image | |
| JP2005345310A (en) | Blood health support system | |
| Dimauro et al. | DeepCilia: Automated, deep-learning based engine for precise ciliary beat frequency estimation | |
| TWI324750B (en) | Microscopic image analysis method and system of microfluidic cells |