CN110287812B - Calculation method of animal circling behavior and application thereof - Google Patents
Calculation method of animal circling behavior and application thereof Download PDFInfo
- Publication number
- CN110287812B CN110287812B CN201910482018.2A CN201910482018A CN110287812B CN 110287812 B CN110287812 B CN 110287812B CN 201910482018 A CN201910482018 A CN 201910482018A CN 110287812 B CN110287812 B CN 110287812B
- Authority
- CN
- China
- Prior art keywords
- animal
- frame
- calculating
- image
- counting
- 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
-
- 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
- G06V40/20—Movements or behaviour, e.g. gesture recognition
Landscapes
- Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the field of video acquisition and image analysis and recognition, in particular to a method for recognizing and calculating the circling behavior of a tested animal driven by specific diseases and medicines by using machine vision. The calculation method of the animal circling behavior comprises the following steps: recording the image; selecting a frame from the image, and counting gray distribution; calculating a threshold value capable of distinguishing an animal from a background, and binarizing each frame of image by taking the threshold value as a boundary; calculating normalized second-order central moment, and obtaining the angle of the animal in each frame to obtain an angle array; and taking a first derivative of the angle array, respectively counting the total number of positive peaks and negative peaks, and counting the number of the positive peaks and the negative peaks, wherein half of the difference value of the total number of the positive peaks and the total number of the negative peaks is the net number of turns of the animal by a method of counting figures after decimal points. The method has small error, accurate data and no interference on animals, and can replace manual counting.
Description
Technical Field
The invention relates to the field of video acquisition and image analysis and recognition, in particular to a method for recognizing and calculating the circling behavior of a tested animal driven by specific diseases and medicines by using machine vision.
Background
About 6 hundred million patients suffering from nervous system diseases such as Parkinson's disease, cerebral ischemia, brain tumor and spinal cord injury are globally present, and with the increasing aging speed of population and the influence of multiple factors such as economy, society and environment, the number of patients suffering from the diseases tends to increase year by year, thus causing huge burden on the development of society and economy.
For many years, neuroscientists and neurologists have been searching for new research tools and treatments for the nervous system including parkinson's disease, cerebral ischemia, brain tumors, and spinal cord injury. In order to study the relationship between neurological disease and animal behavior, it is often necessary to monitor the behavior of model animals under specific conditions. Some specific behaviors can provide quantitative evaluation for pathological characteristics of model animals, such as a unilateral-modeling Parkinson's animal model, and unilateral rotation behaviors presented under the stimulation of specific drugs can directly reflect the degree of specific lesions, so that the circling behaviors of the model animals can be monitored, and quantitative evaluation data can be provided for efficacy evaluation of new drugs and new therapies.
Early animal rotation behavior monitoring mainly comprises methods such as direct observation and manual recording, and the methods have the problems of qualitative subjectivity, time consumption, labor consumption and the like. The existing automatic counting device mostly needs to stick or bind a magnet or stick a color mark on the body surface of an animal, has certain influence on the natural behavior of the animal and brings certain difficulty to experimental operation. Meanwhile, the most common magnet type counter requires a specific magnetic rotor sensing device, and therefore, there are thresholds on the cost and hardware of the device. The device adopting machine vision recognition and video shooting tracking has low hardware requirements, but algorithms adopted by the existing commercial products are based on the calculation of the motion trail of the mass center, and can not accurately recognize the more violent small-axis-distance circling behaviors driven by specific medicines and diseases, for example, when the pathological changes of a unilateral Parkinson's disease model mouse are strong, the unilateral Parkinson's disease model mouse usually presents the in-situ circling behaviors of which the axis distance is smaller than the body axis length and even is zero through medicine stimulation. Therefore, a new trajectory algorithm different from the commonly used behavioral experiments such as open field experiments, maze experiments and the like is urgently needed to be invented, and the algorithm based on the animal body direction is used as a calculation basis so as to realize more accurate revolution behavior counting.
Disclosure of Invention
The invention provides an animal circling behavior calculation method with small error, accurate data and no intervention completely based on machine vision and digital image processing technologies, has less requirements on hardware equipment, can be flexibly configured according to the body types of different experimental animals, can completely identify the small-axis distance circling behavior of a specific disease model animal such as a Parkinson unilateral model mouse, and realizes the flexible, economical, simple, convenient, non-intervention and accurate calculation of the animal circling behavior.
Specifically, the technical scheme of the invention is as follows:
the invention provides a calculation method of animal circling behaviors, which comprises the following steps:
s1: placing animals with constructed disease models in the container and then starting to record images;
s2: selecting a frame from the image, and counting gray distribution;
s3: calculating a threshold value capable of distinguishing an animal from a background, and binarizing each frame of image by taking the threshold value as a boundary;
s4: after the morphology of the animal is filled, calculating a normalized second-order central moment, and obtaining the angle of the animal in each frame to obtain an angle array;
s5: and taking a first derivative of the angle array, respectively counting the total number of positive peaks and negative peaks, and counting the number of the positive peaks and the negative peaks, wherein half of the difference value of the total number of the positive peaks and the total number of the negative peaks is the net number of turns of the animal by a method of counting figures after decimal points.
It should be understood that the present invention is not limited to the above steps, and may also include other steps, such as before step S1, between steps S1 and S2, between steps S2 and S3, between steps S3 and S4, between steps S4 and S5, and after step S5, and other additional steps, without departing from the scope of the present invention.
Preferably, the animal is a mouse.
More preferably, the animal is a mouse in which a model of a neurological disease has been constructed. In a specific embodiment of the invention, the mouse is a parkinsonian unilateral model mouse.
Preferably, in S1, the container is cylindrical and includes a cylindrical main body and a top cover, the cylindrical main body is made of white material, the top cover is made of transparent material, and the top cover is provided with a hole.
The holes are circular holes or irregular holes, in a specific embodiment of the invention, the diameter of the cylinder main body is 25-45cm, the holes on the top cover are circular holes, and the diameter of the holes is 15-25cm.
More preferably, in S1, a monitoring system is connected to the container, and the monitoring system includes a camera fixed to the hole. The monitoring system further comprises a computer. The animal is positioned in the lower visual field of the camera.
Preferably, in S1, images in the container are captured at a sampling rate of 8-15 frames/second for 25-40 minutes.
In an embodiment of the present invention, in S1, the image in the container is captured at a sampling rate of 10 frames/second, and each frame of pixels is 120 × 160 pixels, and the capture time is 30 minutes.
In a preferred embodiment of the present invention, the S2 includes:
s21: randomly selecting a frame, and converting the frame image from a color image into a gray image;
s22: dividing the gray scale into a plurality of groups equally, counting the gray scale of all pixels of the frame according to the groups, and making a columnar distribution map; the x axis of the histogram is a gray value, and the y axis is the number of pixels in a gray range.
It should be understood that the present invention is not limited to the above steps, and may also include other steps, such as before step S21, between steps S21 and S22, and after step S22, and other additional steps, without departing from the protection scope of the present invention.
In one embodiment of the present invention, the gray scales are equally divided into 51 groups from 0 to 255, and each group has an interval of 5.
In a preferred embodiment of the present invention, the S3 includes:
s31: finding two significant peaks on the histogram, setting a minimum value of significance according to the number of pixels of each frame and the size of the animal, and taking a middle value of an interval in which the two peaks are positioned as a threshold value for distinguishing the animal from the background; if there is only one significant peak representing the set of white background pixels, taking half of the value x of the position where the peak is located, namely half of the brightness of the white background, as a threshold value;
s32, performing the following operation on each frame of all the acquired images: newly building result images with the same size, wherein the initial values of all pixels are 0; if the gray value of one pixel point of the frame is lower than the threshold value, the pixel point at the same position on the result image is set as 1, otherwise, the operation is not carried out, the pixel point is kept as 0, and a binary image is obtained.
It should be understood that the present invention is not limited to the above steps, and may also include other steps, such as before step S31, between steps S31 and S32, and after step S32, and other additional steps, without departing from the scope of the present invention.
In an embodiment of the present invention, in S31, according to the number of pixels per frame and the size of the mouse, the degree of significance should be not less than 80, and the middle value of the interval where the two peaks are located is taken as a threshold for distinguishing the black mouse from the white background; if there is only one significant peak representing the set of white background pixels, half of the value x of the position where the peak is located, i.e. half of the brightness of the white background, is taken as a threshold value, so that the white background and the black mouse can be well distinguished.
In a preferred embodiment of the present invention, the S4 includes:
s41: on the binary image, filling treatment is carried out by utilizing the body size of the animal, and noise points are eliminated;
s42: the body axis direction of the animal is obtained by calculating the normalized second-order central moment, the angle of the animal in each frame is obtained, and each frame corresponds to a numerical value, so that an angle array is obtained.
It should be understood that the present invention is not limited to the above steps, and may also include other steps, such as before step S41, between steps S41 and S42, and after step S42, and other additional steps, without departing from the scope of the present invention.
In one embodiment of the present invention, the mouse is filled using the body size of the mouse to eliminate noise caused by whiskers, bare skin, light, and other factors.
In an embodiment of the present invention, the step S42 is as follows: and solving a first derivative of the angle array, namely calculating the change of the central axis angle relative to the previous frame and the next frame in each frame. Every time the axis angle crosses the vertical clockwise, the axis angle changes to +90 degrees in a transient around-90 degrees, so that the first derivative exhibits a positive peak with a value close to 90; when counterclockwise across the vertical, the first derivative exhibits a negative peak, with a value near-90; when rotated between-90 degrees and +90 degrees, the first derivative is typically between +/-35, most of the time being between +/-10. We therefore specify that peak absolute values greater than 40 are the basis for the determination of rotation across the vertical axis.
In a specific embodiment of the present invention, all the times of crossing the vertical direction clockwise and counterclockwise are calculated, and the two values are subtracted to obtain a net value of crossing the vertical direction in a net single direction, wherein a positive value is clockwise and a negative value is counterclockwise. Since the body axis direction does not divide head and tail, this value needs to be divided by two to obtain a net number of turns. For example: the mouse changes the direction of rotation behind 2 times clockwise through vertical direction, and anticlockwise rotation passes through vertical direction, does not reach anticlockwise through numerical value direction once more, changes the direction of rotation once more, and clockwise passes through vertical direction 2 times in succession. The actual net number of turns of the mouse is more than 1 turn when the mouse firstly passes through the vertical direction as a starting point, and the net number of turns given by the method is (2-1 + 2)/2 =1.5 (turn), and the conclusion that the net number of turns is 1 is obtained by a method of not counting figures after the decimal point.
In a second aspect, the invention discloses the use of the above calculation method in the field of neurological diseases.
On the basis of the common general knowledge in the field, the above-mentioned preferred conditions can be combined arbitrarily without departing from the concept and the protection scope of the invention.
Compared with the prior art, the invention has the following remarkable advantages and effects:
(1) The circling behavior of the animal is calculated by monitoring and analyzing the animal image, manual supervision is not needed, and labor and economic cost are saved;
(2) A magnet or a color label is not required to be added on the surface of the animal body, so that a complete non-binding state is realized, and the experiment operation is greatly simplified;
(3) Qualitative subjective judgment is avoided, and the precision is higher;
(4) The animal behavior circling calculation method disclosed by the invention is easy to configure according to the body sizes of different animals;
(5) The animal behavior circling calculation method disclosed by the invention is particularly accurate in identification of small-wheelbase circling behaviors, has an obvious effect, and can accurately replace manual counting;
(6) The method disclosed by the invention can overcome individual differences of animals, is not influenced by actions of standing, jumping and the like of the animals in an experiment, and is insensitive to image resolution difference caused by focal length change of the camera.
Drawings
FIG. 1 is an exploded view of hardware components in an embodiment of the invention;
FIG. 2 is an assembly diagram of hardware components in an embodiment of the invention;
FIG. 3 is a flow chart of a method in an embodiment of the present invention;
FIG. 4 is a schematic diagram of S2-S4 in an embodiment of the present invention;
FIG. 5 is a schematic diagram illustrating the calculation of the axial angle value according to an embodiment of the present invention;
FIG. 6 is a schematic diagram illustrating the calculation of the first derivative according to an embodiment of the present invention;
in the figure:
1-cylinder main body, 2-top cover, 3-camera and 4-computer.
Detailed Description
The technical solutions of the present invention are described in detail below with reference to the drawings and embodiments, but the present invention is not limited to the scope of the embodiments.
Experimental procedures without specifying specific conditions in the following examples were selected in accordance with conventional procedures and conditions, or in accordance with commercial instructions. The reagents and starting materials used in the present invention are commercially available.
Example 1
The embodiment discloses a calculation method of mouse circling behavior, which comprises the following steps:
s1: placing a Parkinson's disease unilateral model mouse in a container and then starting to record an image;
s2: selecting a frame from the image, and counting gray distribution;
s3: calculating a threshold value capable of distinguishing a mouse from a background, and binarizing each frame of image by taking the threshold value as a boundary;
s4: after the morphology of the animal is filled, calculating a normalized second-order central moment, and obtaining the angle of the mouse in each frame to obtain an angle array;
s5: and taking a first derivative of the angle array, respectively counting the total number of positive peaks and negative peaks, and counting the number of the positive peaks and the negative peaks, wherein half of the difference value of the total number of the positive peaks and the total number of the negative peaks is the net number of turns of the animal by a method of counting figures after decimal points.
Specifically, an explosion diagram and an assembly diagram of hardware parts in the method are respectively shown in fig. 1 and fig. 2. The method flow chart in this embodiment is shown in fig. 3. The hardware portion includes a container and a monitoring system. The container is cylindrical and comprises a cylindrical main body 1 and a top cover 2, wherein the cylindrical main body 1 is made of white materials, the top cover 2 is made of transparent materials, and holes are formed in the top cover 2. The diameter of the cylinder main body is 25-45cm, the hole in the top cover is a circular hole, the diameter of the hole is 15-25cm, and the hole is circular.
The monitoring system comprises a camera 3, and the camera 3 is fixed on the hole. The monitoring system further comprises a computer 4. The mouse is positioned in the visual field below the camera.
In S1, the image in the container is captured at a sampling rate of 10 frames/second, with pixels of 120x160 pixels per frame, acquired for 30 minutes.
The S2 comprises the following steps:
s21: randomly selecting a frame, and converting the frame image from a color image into a gray image, as shown in fig. 4 (a);
s22: dividing the gray scale into a plurality of groups equally, counting the gray scale of all pixels of the frame according to the groups, and making a columnar distribution map; the x-axis of the histogram is a gray value, the y-axis is the number of pixels in a gray range, the gray is equally divided into 51 groups from 0 to 255, and each group has an interval of 5, as shown in fig. 4 (b).
The S3 comprises the following steps:
s31: finding two significant peaks on the columnar distribution map, setting the significance to be not less than 80 according to the number of pixels of each frame and the size of the animal, and taking the middle value of the interval of the two peaks as a threshold value for distinguishing the animal from the background; if there is only one significant peak representing the set of white background pixels, taking half of the value x of the position where the peak is located, i.e. half of the brightness of the white background, as a threshold, the white background can be well distinguished from the black mouse, as shown in fig. 4 (c);
s32, performing the following operation on each frame of all the acquired images: newly building result images with the same size, wherein the initial values of all pixels are 0; if the gray value of one pixel point of the frame is lower than the threshold, the pixel point at the same position on the result image is set as 1, otherwise, the operation is not performed, the pixel point is kept as 0, and a binary image is obtained, as shown in fig. 4 (d).
The S4 comprises the following steps:
s41: on the binary image, filling processing is carried out by utilizing the body size of the animal, and noise points are eliminated, as shown in fig. 4 (e);
s42: the body axis direction of the animal is obtained by calculating the normalized second-order central moment, the angle of the animal in each frame is obtained, and each frame corresponds to a numerical value, so that an angle array is obtained.
S42, the specific steps are as follows: the body axis direction of the mouse is obtained by calculating the normalized second-order central moment, and is an angle value without dividing the head and the tail, the upper left and the lower right direction is a negative value, and the upper right and the lower left direction is a positive value, as shown in fig. 5. Each frame corresponds to a numerical value, and an angle array is obtained.
And (4) solving a first derivative of the angle array, namely calculating the change of the central axis angle relative to the previous frame and the next frame in each frame. Every time the axis angle crosses the vertical clockwise, the axis angle changes to +90 degrees in a transient around-90 degrees, so that the first derivative exhibits a positive peak with a value close to 90; when counterclockwise across the vertical, the first derivative exhibits a negative peak, with a value near-90; when rotated between-90 degrees and +90 degrees, the first derivative is typically between +/-35, most of the time being between +/-10. We therefore specify that peak absolute values greater than 40 are the basis for the determination of rotation across the vertical axis, as shown in fig. 6.
In a specific embodiment of the present invention, all the times of crossing the vertical direction clockwise and counterclockwise are calculated, and the two values are subtracted to obtain a net value of crossing the vertical direction in a net single direction, wherein a positive value is clockwise and a negative value is counterclockwise. Since the body axis direction does not divide head and tail, this value needs to be divided by two to obtain a net number of turns. Therefore, by counting the number after the decimal point, half of the difference between the total number of the positive peaks and the total number of the negative peaks is the net number of turns of the animal.
For example: the mouse changes the direction of rotation behind 2 times through vertical direction clockwise, and anticlockwise rotation passes through vertical direction, does not reach anticlockwise through vertical direction once more, changes the direction of rotation once more, clockwise continuous through vertical direction 2 times. The actual net rotation number of the mouse is more than 1 circle when the mouse firstly passes through the vertical direction, and the net rotation number given by the method is (2-1 + 2)/2 =1.5 (circle), and the conclusion that the net rotation number is 1 is obtained by a method of not counting figures after a decimal point.
The method disclosed by the embodiment has the following advantages: (1) The circling behavior of the animal is calculated by monitoring and analyzing the animal image, manual supervision is not needed, and labor and economic cost are saved; (2) A magnet or a color label is not required to be added on the surface of the animal body, so that a complete non-binding state is realized, and the experiment operation is greatly simplified; (3) qualitative subjective judgment is avoided, and the precision is higher; (4) The animal behavior circling calculation method disclosed by the invention is easy to configure according to the body sizes of different animals; (5) The method is particularly accurate in identification of small-wheelbase turning behavior, obvious in effect and capable of accurately replacing manual counting; (6) The method can overcome individual differences of animals, is not influenced by actions of standing, jumping and the like of the animals in experiments, and is insensitive to image resolution difference caused by focal length change of the camera.
Example 2
In this example, the same video file was compared by manual counting and using the algorithm disclosed in example 1 for 30 minutes, and the test conditions were identical except for the number of turns calculated. The results of the tests are shown in Table 1, comparing the difference between the net number of revolutions from the automatic and manual counts with a t-test in pairs, giving a p-value of 0.2956, which is considered as no statistical difference. Looking again at the specific numerical differences, the maximum absolute value of the difference was 2, errors in manual counting were not excluded, and the difference was represented as 0.06 cycles/min when calculating the number of net revolutions per minute, and was not considered as a significant difference in animal experiments. Therefore, the algorithm can be used as an accurate substitute for manual counting in animal rotation experiments.
TABLE 1 number of revolutions in 30 minutes for the same video file by manual counting and automatic counting according to the algorithm
The above embodiments are preferred embodiments of the present invention, but the present invention is not limited to the above embodiments, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be construed as equivalents thereof, and all such changes, modifications, substitutions, combinations, and simplifications are intended to be included in the scope of the present invention.
Claims (11)
1. A calculation method for animal circling behaviors is characterized by comprising the following steps:
s1: placing animals with constructed disease models in the container and then starting to record images;
s2: selecting a frame from the image, and counting gray distribution;
s3: calculating a threshold value capable of distinguishing an animal from a background, and binarizing each frame of image by taking the threshold value as a boundary;
s4: after the morphology of the animal is filled, the angle of the animal in each frame is obtained by calculating the body axis direction of the animal, and an angle array is obtained;
s5: taking a first derivative of the angle array, respectively counting the total number of positive peaks and negative peaks, and counting the number of the positive peaks and the negative peaks, wherein half of the difference between the total number of the positive peaks and the total number of the negative peaks is the net number of turns of the animal by a method of counting figures after decimal points;
in S5, the total number of the positive peaks is that every time the axis angle clockwise spans the vertical direction, the first derivative presents a positive peak value, and all times of clockwise spanning the vertical direction are calculated;
in the step S5, the total number of the negative peaks is that when the counter-clockwise crosses the vertical direction, the first derivative presents a negative peak value, and all times of counter-clockwise crossing the vertical direction are calculated.
2. The method for calculating circling behavior of animal according to claim 1, wherein the animal is a mouse.
3. The method for calculating circling behavior of animal according to claim 2, wherein the animal is a mouse with a model of neurological disease.
4. The method for calculating animal circling behavior according to claim 1, characterized in that in S1, the container is cylindrical and comprises a cylindrical main body and a top cover, wherein the cylindrical main body is made of white material, the top cover is made of transparent material, and holes are arranged on the top cover.
5. The method of claim 4, wherein in S1, a monitoring system is connected to the container, the monitoring system comprises a camera, and the camera is fixed to the hole.
6. The method of claim 1, wherein the images in the container are captured at a sampling rate of 8-15 frames/second for 25-40 minutes in S1.
7. The method for calculating animal circling behavior according to claim 1, wherein the S2 comprises:
s21: randomly selecting a frame, and converting the frame image from a color image into a gray image;
s22: dividing the gray level into a plurality of groups equally, counting the gray levels of all pixels of the frame according to the groups, and making a columnar distribution map; the x axis of the histogram is a gray value, and the y axis is the number of pixels in a gray range.
8. The method of calculating animal circling behavior according to claim 7, characterized in that in S22, the gradation is equally divided into 51 groups by 0 to 255, each group interval being 5.
9. The method for calculating animal circling behavior according to claim 7, wherein the S3 comprises:
s31: finding two significant peaks on the histogram, setting a minimum value of significance according to the number of pixels of each frame and the size of the animal, and taking a middle value of an interval in which the two peaks are positioned as a threshold value for distinguishing the animal from the background; if there is only one significant peak representing the set of white background pixels, taking half of the value x of the position where the peak is located, namely half of the brightness of the white background, as a threshold value;
s32, performing the following operation on each frame of all the acquired images: newly building result images with the same size, wherein the initial values of all pixels are 0; if the gray value of one pixel point of the frame is lower than the threshold value, the pixel point at the same position on the result image is set as 1, otherwise, the operation is not carried out, the pixel point is kept as 0, and a binary image is obtained.
10. The method for calculating animal circling behavior according to claim 9, wherein the S4 comprises:
s41: on the binary image, filling treatment is carried out by utilizing the body size of the animal, and noise points are eliminated;
s42: the body axis direction of the animal is obtained by calculating the normalized second-order central moment, the angle of the animal in each frame is obtained, and each frame corresponds to a numerical value, so that an angle array is obtained.
11. Use of the calculation method according to any one of claims 1 to 10 in the field of neurological diseases.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910482018.2A CN110287812B (en) | 2019-06-04 | 2019-06-04 | Calculation method of animal circling behavior and application thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910482018.2A CN110287812B (en) | 2019-06-04 | 2019-06-04 | Calculation method of animal circling behavior and application thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110287812A CN110287812A (en) | 2019-09-27 |
| CN110287812B true CN110287812B (en) | 2023-04-07 |
Family
ID=68003319
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910482018.2A Active CN110287812B (en) | 2019-06-04 | 2019-06-04 | Calculation method of animal circling behavior and application thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110287812B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111727905A (en) * | 2020-06-22 | 2020-10-02 | 安徽正华生物仪器设备有限公司 | Automatic analysis system and method for big and small mouse open field experiment based on deep learning |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004021282A1 (en) * | 2002-08-29 | 2004-03-11 | Japan Science And Technology Agency | Animal behavior analysis method, animal behavior analysis system, animal behavior analysis program, and computer-readable recorded medium on which the program is recorded |
| JP2005196398A (en) * | 2004-01-06 | 2005-07-21 | Ohara Ika Sangyo Kk | Small animal behavior analysis device, small animal behavior analysis method, and behavior diagnosis method |
| CN107025652A (en) * | 2017-05-05 | 2017-08-08 | 太原理工大学 | A kind of flame detecting method based on kinetic characteristic and color space time information |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7496228B2 (en) * | 2003-06-13 | 2009-02-24 | Landwehr Val R | Method and system for detecting and classifying objects in images, such as insects and other arthropods |
-
2019
- 2019-06-04 CN CN201910482018.2A patent/CN110287812B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004021282A1 (en) * | 2002-08-29 | 2004-03-11 | Japan Science And Technology Agency | Animal behavior analysis method, animal behavior analysis system, animal behavior analysis program, and computer-readable recorded medium on which the program is recorded |
| JP2005196398A (en) * | 2004-01-06 | 2005-07-21 | Ohara Ika Sangyo Kk | Small animal behavior analysis device, small animal behavior analysis method, and behavior diagnosis method |
| CN107025652A (en) * | 2017-05-05 | 2017-08-08 | 太原理工大学 | A kind of flame detecting method based on kinetic characteristic and color space time information |
Non-Patent Citations (2)
| Title |
|---|
| 基于USB摄像头的通用Morris水迷宫实验系统;刘渊等;《中国医疗设备》;20090515(第05期);全文 * |
| 基于计算机视觉的鱼苗自动计数系统研究;王文静等;《渔业现代化》;20160620(第03期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110287812A (en) | 2019-09-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Gillebert et al. | Automated delineation of stroke lesions using brain CT images | |
| Serences et al. | Estimating the influence of attention on population codes in human visual cortex using voxel-based tuning functions | |
| US20220139543A1 (en) | Method and system for seizure detection | |
| Epstein et al. | Neural responses to visual scenes reveals inconsistencies between fMRI adaptation and multivoxel pattern analysis | |
| Dong et al. | Minian, an open-source miniscope analysis pipeline | |
| KR102321601B1 (en) | Biological Classification Devices and Methods for Alzheimer's Disease Using Multimodal Brain Image | |
| BR112013004642B1 (en) | MONITORING OR PREDICTION SYSTEM TO DETECT THE START OF A NEUROLOGICAL EPISODE AND METHOD TO DETECT THE START OF A NEUROLOGICAL EPISODE | |
| US20170065229A1 (en) | Neural oscillation monitoring system | |
| EP3761871A1 (en) | Method for early prediction of neurodegenerative decline | |
| CN110840468A (en) | Autism risk assessment method and device, terminal device and storage medium | |
| CN110287812B (en) | Calculation method of animal circling behavior and application thereof | |
| CN1359656A (en) | Method and device for estimating neure damage degree in cerabral cortex | |
| JP6590690B2 (en) | Cell image retrieval apparatus and method, and program | |
| CN112674782B (en) | Device and method for detecting epileptic-like electrical activity of epileptic during inter-seizure period | |
| Thakral et al. | The hippocampus is sensitive to the mismatch in novelty between items and their contexts | |
| CN104867153A (en) | Detection system based on phosphorylated tau protein content information in brain magnetic resonance image | |
| CN110298239A (en) | Target monitoring method, apparatus, computer equipment and storage medium | |
| WO2020152953A1 (en) | Discerning device, cell mass discerning method, and computer program | |
| CN113143275B (en) | Electroencephalogram fatigue detection method for quantitative evaluation of sample and characteristic quality in combined manner | |
| CN114999657A (en) | Method for evaluating the association of nerve fiber bundles and gait disturbances in patients with Parkinson's disease and related products | |
| Kay | Understanding visual representation by developing receptive-field models | |
| Geman et al. | Partitioning methods used in DBS treatments analysis results | |
| EP2991035A1 (en) | Method and device for representing pathological changes within an examined object based on 3d data sets | |
| CN115429291B (en) | Epileptic detection system combining multiscale scatter entropy and multi-label K nearest neighbor | |
| CN110652307B (en) | Functional nuclear magnetic image-based striatum function detection system for schizophrenia patient |
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 |