CN104881636A - Method and device for identifying lying behavior of pig - Google Patents
Method and device for identifying lying behavior of pig Download PDFInfo
- Publication number
- CN104881636A CN104881636A CN201510232700.8A CN201510232700A CN104881636A CN 104881636 A CN104881636 A CN 104881636A CN 201510232700 A CN201510232700 A CN 201510232700A CN 104881636 A CN104881636 A CN 104881636A
- Authority
- CN
- China
- Prior art keywords
- pig
- pixel
- identified
- height
- ground
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/103—Static body considered as a whole, e.g. static pedestrian or occupant recognition
Landscapes
- Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method and a device for identifying a lying behavior of a pig. The method and the device may improve accuracy of pig lying behavior identification. The method comprises: acquiring a binary image and a depth image of a pig to be identified; fitting the binary image to an elliptical figure, computing the horizontal coordinate of the left end point and the horizontal coordinate of the right end point of the long axis of the elliptical figure, and the coordinate of the central point of the elliptical figure; computing the height, above the ground, of each pixel point on a line segment in the depth image, wherein the line segment passes through the coordinate of the central point and is in parallel with an X axis, and the horizontal coordinate of the left end point and the horizontal coordinate of the right end point of the line segment are equal to those of the long axis respectively; and comparing a maximum value in the heights, above the ground, of all the pixel points with a preset value, and determining that the pig to be identified is lying if the maximum value is less than the preset value.
Description
Technical field
The present invention relates to computer image technology field, be specifically related to a kind of method and the device that identify pig lying behaviour.
Background technology
Research shows, the time pig of about 75% is the state of couching, and therefore, the comfortable shared area that couches is an important references of pig house area and quality thereof.In addition, lying behaviour is also that pig carries out a thermotaxic important means (such as, changing position and lying period), and lying behaviour is directly related with environment temperature.And for piglet, the change suddenly of sow behavior likely can extrude or trample piggy, thus the survival rate of the front piggy of impact wean.Thus, the lying behaviour of sow is not only closely bound up with the welfare of piggy, the reference of sow farrowing bed middle thermolamp placement location and pig house size design is can be used as by the lying behaviour studying sow, and likely reduce the probability of sow extruding piggy, improve porkling survival rate and last production efficiency.
Mostly depended on the experience level of cultivation expert in tradition research by the process of manual observation pig lying behaviour, there is the problem that subjectivity is strong, range of observation is limited, cost of labor is high, accuracy is not high, mistake is difficult to avoid, cannot realize the deficiencies such as long-term Real-Time Monitoring.At present, utilize image processing techniques automatically identify pig behavior have leave alone pig, in real time, continuously, objective advantage, become new study hotspot.Digital image processing techniques are current pig behavior automatic identification technology means the most conventional, can be used for the horizontal direction properties of distributions identifying pig.And the lying behaviour of pig is except the feature having horizontal distribution, vertical height also exists larger difference, therefore, existing digital image processing techniques can not identify the lying behaviour of pig exactly.
Summary of the invention
The object of the invention is to, a kind of method and the device that identify pig lying behaviour are provided, the accuracy of pig lying behaviour identification can be improved.
For this purpose, on the one hand, the present invention proposes a kind of method identifying pig lying behaviour, comprising:
Obtain bianry image and the depth image of pig to be identified;
Described bianry image is fitted to ellipse figure, and calculates the horizontal ordinate of the left end point of the major axis of described ellipse figure, the horizontal ordinate of right endpoint, the center point coordinate of described ellipse figure;
Calculate in described depth image cross described center point coordinate parallel with X-axis and on the line segment that the horizontal ordinate of left end point and right endpoint is equal with the horizontal ordinate of right endpoint with the left end point of described major axis respectively each pixel to the height on ground;
Described pixel being compared to the maximal value in the height on ground and the numerical value preset, if described maximal value is less than described default numerical value, determining that the behavior of described pig to be identified is for couching.
On the other hand, the present invention proposes a kind of device identifying pig lying behaviour, comprising:
Image acquisition unit, for obtaining bianry image and the depth image of pig to be identified;
Coordinate calculating unit, for described bianry image is fitted to ellipse figure, and calculates the horizontal ordinate of the left end point of the major axis of described ellipse figure, the horizontal ordinate of right endpoint, the center point coordinate of described ellipse figure;
Height calculation unit, for calculate in described depth image cross described center point coordinate parallel with X-axis and on the line segment that the horizontal ordinate of left end point and right endpoint is equal with the horizontal ordinate of right endpoint with the left end point of described major axis respectively each pixel to the height on ground;
Recognition unit, for being compared to the maximal value in the height on ground and the numerical value preset by described pixel, if described maximal value is less than described default numerical value, determines that the behavior of described pig to be identified is for couching.
The method of the identification pig lying behaviour described in the embodiment of the present invention and device, the horizontal ordinate of the left end point of the major axis of the ellipse figure that the bianry image that both make use of pig to be identified fits to, the horizontal ordinate of right endpoint, and center point coordinate, make use of again in the depth image of pig to be identified and cross described center point coordinate parallel with X-axis and on the line segment that the horizontal ordinate of left end point and right endpoint is equal with the horizontal ordinate of right endpoint with the left end point of described major axis respectively, each pixel compares to the maximal value in the height on ground and the numerical value preset, determine the lying behaviour of pig to be identified, namely make use of the horizontal direction properties of distributions of pig to be identified and vertical direction properties of distributions to identify lying behaviour, thus the horizontal direction properties of distributions of pig is only utilized than existing recognition technology, the vertical direction properties of distributions of pig is not utilized to identify, the accuracy of pig lying behaviour identification can be improved.
Accompanying drawing explanation
Fig. 1 is the schematic flow sheet of method one embodiment of identification pig lying behaviour of the present invention;
In Fig. 2, (a) and (b) are respectively the second digital picture in another embodiment of method of identification pig lying behaviour of the present invention and bianry image;
Fig. 3 is the frame structure schematic diagram of device one embodiment of identification pig lying behaviour of the present invention.
Embodiment
For making the object of the embodiment of the present invention, technical scheme and advantage clearly, below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly described, obviously, described embodiment is the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.
As shown in Figure 1, the present embodiment discloses a kind of method identifying pig lying behaviour, comprising:
S1, the bianry image obtaining pig to be identified and depth image;
S2, described bianry image is fitted to ellipse figure, and calculate the horizontal ordinate of the left end point of the major axis of described ellipse figure, the horizontal ordinate of right endpoint, the center point coordinate of described ellipse figure;
S3, calculate in described depth image cross described center point coordinate parallel with X-axis and on the line segment that the horizontal ordinate of left end point and right endpoint is equal with the horizontal ordinate of right endpoint with the left end point of described major axis respectively each pixel to the height on ground;
S4, by described pixel to the maximal value in the height on ground from preset numerical value (kind of pig to be identified is different, this numerical value has corresponding difference, if pig to be identified is Landrace, then this numerical value can be 50cm) compare, if described maximal value is less than described default numerical value, determine that the behavior of described pig to be identified is for couching.
The depth image uncorrected data of original acquisition is text data, and each depth image packet contains 640 × 480 numerical value, and each numerical value is the binary data of 16.
The algorithm involved by method of the identification pig lying behaviour described in the embodiment of the present invention can under MATLAB R2010a environment programming realization.
The method of the identification pig lying behaviour described in the embodiment of the present invention, the horizontal ordinate of the left end point of the major axis of the ellipse figure that the bianry image that both make use of pig to be identified fits to, the horizontal ordinate of right endpoint, and center point coordinate, make use of again in the depth image of pig to be identified and cross described center point coordinate parallel with X-axis and on the line segment that the horizontal ordinate of left end point and right endpoint is equal with the horizontal ordinate of right endpoint with the left end point of described major axis respectively, each pixel compares to the maximal value in the height on ground and the numerical value preset, determine the lying behaviour of pig to be identified, namely make use of the horizontal direction properties of distributions of pig to be identified and vertical direction properties of distributions to identify lying behaviour, thus the horizontal direction properties of distributions of pig is only utilized than existing recognition technology, the vertical direction properties of distributions of pig is not utilized to identify, the accuracy of pig lying behaviour identification can be improved.
Alternatively, as shown in (a) He (b) in Fig. 2, in another embodiment of the method for identification pig lying behaviour of the present invention, the bianry image of described acquisition pig to be identified, comprising:
Obtain the first digital picture under feeding environment and the second digital picture; Wherein, described first digital picture does not comprise pig to be identified, and described second digital picture comprises described pig to be identified;
By described second digital picture is deducted described first digital picture, obtain the digital picture of described pig to be identified, and the digital picture of described pig to be identified is carried out two-value conversion, obtain the bianry image of described pig to be identified.
In the embodiment of the present invention, the first digital picture under feeding environment, second digital picture and depth image can obtain by being positioned over the video camera shooting of photograph under the high perpendicular of pig house 2.2 meters, positive middle distance ground, first digital picture can take acquisition before pig house does not also have pig to be identified, or by having the image of pig to reduce and splicing acquisition (because some region does not have pig in some image shot by camera, in other image, other region does not have pig, these be there is no the region merging technique of pig, the complete background image not having pig can be obtained).Consider actual scene, can after acquisition first digital picture, first by pig to be identified in the first digital picture the region that hides by top pipeline restore, carry out subsequent treatment more afterwards, or can after the image being photographed pig to be identified by video camera, first by have pig to be identified in the image of pig to be identified the region that hides by top pipeline restore, carry out subsequent treatment more afterwards.After the first digital picture obtained under feeding environment and the second digital picture, consider noise and subsequent treatment speed, the subregion (comprising hopper, wall etc.) of the first digital picture and the second digital picture surrounding can be removed.After the digital picture obtaining pig to be identified, the digital picture of pig to be identified can be strengthened by applied morphology filtering method.After the digital picture of pig to be identified is converted to bianry image, considers actual scene, holes filling can be carried out to bianry image, to obtain the bianry image of smooth pig to be identified.As shown in Figure 2, in Fig. 2, (a) is the second digital picture, and in Fig. 2, (b) is the bianry image of pig in (a) of Fig. 2.
Alternatively, in another embodiment of the method for identification pig lying behaviour of the present invention, described described bianry image is fitted to ellipse figure, comprising:
By minimum two-value matching, described bianry image is fitted to ellipse figure.
Alternatively, in another embodiment of the method for identification pig lying behaviour of the present invention, described center point coordinate parallel with X-axis is crossed and on the line segment that the horizontal ordinate of left end point and right endpoint is equal with the horizontal ordinate of right endpoint with the left end point of described major axis respectively, each pixel, to the height on ground, comprising in the described depth image of described calculating:
For each pixel on described line segment, this pixel calculated and straight line vertical with described line segment is the depth value of each pixel of 0,1 and 2 with the pixel distance of this pixel, and to calculate described according to formula h eight=h-d be that the pixel of 0,1 and 2 is to the height on ground with the pixel distance of this pixel, described height is the height of pixel to ground, described h is the height of camera to ground, and described d is the depth value of pixel;
For each pixel on described line segment, this pixel calculated and the straight line vertical with described line segment with the pixel distance of this pixel is the pixel of 0,1 and 2 to the average of the height on ground, and using this average as this pixel to the height on ground.
In the embodiment of the present invention, can be moved to right 3 by high 13 bit value of the naked depth data by pixel and calculate the depth value of pixel.Such as, the naked depth data of depth image pixel is: 00011100011011000, its naked pixel value=14552, is moved to right by naked depth data 3, then this pixel is 0000101000111011 to the range data of camera, i.e. depth value=the 1819mm of this pixel.
Alternatively, in another embodiment of the method for identification pig lying behaviour of the present invention, the horizontal ordinate of the left end point of the major axis of the described ellipse figure of described calculating, the horizontal ordinate of right endpoint, the center point coordinate of described ellipse figure, also comprises:
Calculate the major axis of described ellipse figure and the angle of X-axis;
Determining that the behavior of described pig to be identified is for after couching, according to the horizontal ordinate of the left end point of described major axis, the horizontal ordinate of right endpoint, the direction that couches of pig to be identified described in the center point coordinate of described ellipse figure and described angle identification.
Alternatively, in another embodiment of the method for identification pig lying behaviour of the present invention, the horizontal ordinate of the described left end point according to described major axis, the horizontal ordinate of right endpoint, the direction that couches of pig to be identified described in the center point coordinate of described ellipse figure and described angle identification, comprising:
Pig to be identified with the standard deviation of all point of the described major axis Zhong Mei side, both sides that is separator bar to the distance of described major axis according to the center point coordinate of described ellipse figure and described angle calcu-lation;
Described standard deviation is compared, if be greater than back to all point of thermolamp side to the standard deviation of the distance of described major axis towards all point of thermolamp side to the standard deviation of the distance of described major axis, then determine that the direction that couches of described pig to be identified is that breast side couches towards thermolamp, otherwise, determine that the direction that couches of described pig to be identified is that dorsal part couches towards thermolamp.
In the embodiment of the present invention, pig breast side does not have dorsal part smooth, thus on breast side each pixel to long wheelbase from standard deviation be greater than each pixel on dorsal part to long wheelbase from standard deviation.But, in time having piggy to follow head or the afterbody adhesion of sow, then both may be caused to change by standard extent, cause erroneous judgement.For avoiding the generation of this kind of situation, the horizontal ordinate scope of the point of pig to be identified can be adjusted to horizontal ordinate+100 pixels from horizontal ordinate-100 pixels of the left end point of major axis to the right endpoint of major axis.
Present invention achieves the automatic identification pig lying behaviour of a kind of fusing digital images and depth image and the method in the direction that couches thereof, can be used as the reference improving pig house size and thermolamp placement location, and pig welfare and production efficiency can be improved.Can (acquisition interval can be arranged as required by using economical and practical 3D picture pick-up device (as the KinectTM for Windows that Microsoft company produces) carry out the synchronous acquisition of depth image and digital picture, such as 5s gathers once or 10s gathers once), and fusion treatment and the algorithm development of the degree of depth and digital picture is carried out based on Matlab, handled result is preserved into database, and the image of process gained is then preserved into corresponding catalogue according to the date.The result processed the view data of 3D picture pick-up device four days shooting is as shown in the horizontal data in table 1 and table 2:
Table 1
Table 2
Table 1 is the recognition result of pig lying behaviour to be identified, and in table 1, the result of artificial cognition is longitudinal data, and the result of algorithm identification is horizontal data.Algorithm has processed the 14290 width degree of depth and 14290 corresponding width digital pictures respectively, and algorithm correctly can identify 11936 lying behaviour, and missing 2 other Activity recognition is lying behaviour, and the recognition correct rate of lying behaviour is 99.98%; 9 lying behaviour are identified as other behavior by algorithm by mistake, and 2343 other behaviors can be correctly validated out.
Table 2 to couch the recognition result in direction for pig to be identified, and in table 2, the result of artificial cognition is longitudinal data, and the result of algorithm identification is horizontal data.Always have 5678 breast sides in algorithm identification to be correctly validated towards the lying behaviour of thermolamp, but have 135 dorsal parts to be erroneously identified as breast side towards thermolamp towards the lying behaviour of thermolamp, algorithm accuracy is 97.68%.And for dorsal part towards in the lying behaviour identification of thermolamp, algorithm correctly identifies 5949, and 175 are mistaken for breast side towards thermolamp, and accuracy rate reaches 97.14%.The discrimination in direction of always couching is 97.40%.
As shown in Figure 3, the present embodiment discloses a kind of device identifying pig lying behaviour, comprising:
Image acquisition unit 1, for obtaining bianry image and the depth image of pig to be identified;
Coordinate calculating unit 2, for described bianry image is fitted to ellipse figure, and calculates the horizontal ordinate of the left end point of the major axis of described ellipse figure, the horizontal ordinate of right endpoint, the center point coordinate of described ellipse figure;
Height calculation unit 3, for calculate in described depth image cross described center point coordinate parallel with X-axis and on the line segment that the horizontal ordinate of left end point and right endpoint is equal with the horizontal ordinate of right endpoint with the left end point of described major axis respectively each pixel to the height on ground;
Recognition unit 4, for being compared to the maximal value in the height on ground and the numerical value preset by described pixel, if described maximal value is less than described default numerical value, determines that the behavior of described pig to be identified is for couching.
The device of the identification pig lying behaviour described in the embodiment of the present invention, the horizontal ordinate of the left end point of the major axis of the ellipse figure that the bianry image that both make use of pig to be identified fits to, the horizontal ordinate of right endpoint, and center point coordinate, make use of again in the depth image of pig to be identified and cross described center point coordinate parallel with X-axis and on the line segment that the horizontal ordinate of left end point and right endpoint is equal with the horizontal ordinate of right endpoint with the left end point of described major axis respectively, each pixel compares to the maximal value in the height on ground and the numerical value preset, determine the lying behaviour of pig to be identified, namely make use of the horizontal direction properties of distributions of pig to be identified and vertical direction properties of distributions to identify lying behaviour, thus the horizontal direction properties of distributions of pig is only utilized than existing recognition technology, the vertical direction properties of distributions of pig is not utilized to identify, the accuracy of pig lying behaviour identification can be improved.
Alternatively, in another embodiment of the device of identification pig lying behaviour of the present invention, described image acquisition unit, comprising:
First Image Acquisition subelement, for obtaining the first digital picture under feeding environment and the second digital picture; Wherein, described first digital picture does not comprise pig to be identified, and described second digital picture comprises described pig to be identified;
Second Image Acquisition subelement, for by described second digital picture is deducted described first digital picture, obtain the digital picture of described pig to be identified, and the digital picture of described pig to be identified is carried out two-value conversion, obtain the bianry image of described pig to be identified.
Alternatively, in another embodiment of the device of identification pig lying behaviour of the present invention, described coordinate calculating unit, for by minimum two-value matching, fits to ellipse figure by described bianry image.
Alternatively, in another embodiment of the device of identification pig lying behaviour of the present invention, described height calculation unit, comprising:
First high computational subelement, for for each pixel on described line segment, this pixel calculated and straight line vertical with described line segment is the depth value of each pixel of 0,1 and 2 with the pixel distance of this pixel, and to calculate described according to formula h eight=h-d be that the pixel of 0,1 and 2 is to the height on ground with the pixel distance of this pixel, described height is the height of pixel to ground, described h is the height of camera to ground, and described d is the depth value of pixel;
Second high computational subelement, for for each pixel on described line segment, this pixel calculated and the straight line vertical with described line segment with the pixel distance of this pixel is the pixel of 0,1 and 2 to the average of the height on ground, and using this average as this pixel to the height on ground.
Although describe embodiments of the present invention by reference to the accompanying drawings, but those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present invention, such amendment and modification all fall into by within claims limited range.
Claims (10)
1. identify a method for pig lying behaviour, it is characterized in that, comprising:
Obtain bianry image and the depth image of pig to be identified;
Described bianry image is fitted to ellipse figure, and calculates the horizontal ordinate of the left end point of the major axis of described ellipse figure, the horizontal ordinate of right endpoint, the center point coordinate of described ellipse figure;
Calculate in described depth image cross described center point coordinate parallel with X-axis and on the line segment that the horizontal ordinate of left end point and right endpoint is equal with the horizontal ordinate of right endpoint with the left end point of described major axis respectively each pixel to the height on ground;
Described pixel being compared to the maximal value in the height on ground and the numerical value preset, if described maximal value is less than described default numerical value, determining that the behavior of described pig to be identified is for couching.
2. the method for identification pig lying behaviour according to claim 1, it is characterized in that, the bianry image of described acquisition pig to be identified, comprising:
Obtain the first digital picture under feeding environment and the second digital picture; Wherein, described first digital picture does not comprise pig to be identified, and described second digital picture comprises described pig to be identified;
By described second digital picture is deducted described first digital picture, obtain the digital picture of described pig to be identified, and the digital picture of described pig to be identified is carried out two-value conversion, obtain the bianry image of described pig to be identified.
3. the method for identification pig lying behaviour according to claim 1, is characterized in that, described described bianry image is fitted to ellipse figure, comprising:
By minimum two-value matching, described bianry image is fitted to ellipse figure.
4. the method for identification pig lying behaviour according to claim 1, it is characterized in that, described center point coordinate parallel with X-axis is crossed and on the line segment that the horizontal ordinate of left end point and right endpoint is equal with the horizontal ordinate of right endpoint with the left end point of described major axis respectively, each pixel, to the height on ground, comprising in the described depth image of described calculating:
For each pixel on described line segment, this pixel calculated and straight line vertical with described line segment is the depth value of each pixel of 0,1 and 2 with the pixel distance of this pixel, and to calculate described according to formula h eight=h-d be that the pixel of 0,1 and 2 is to the height on ground with the pixel distance of this pixel, described height is the height of pixel to ground, described h is the height of camera to ground, and described d is the depth value of pixel;
For each pixel on described line segment, this pixel calculated and the straight line vertical with described line segment with the pixel distance of this pixel is the pixel of 0,1 and 2 to the average of the height on ground, and using this average as this pixel to the height on ground.
5. the method for identification pig lying behaviour according to claim 1, it is characterized in that, the horizontal ordinate of the left end point of the major axis of the described ellipse figure of described calculating, the horizontal ordinate of right endpoint, the center point coordinate of described ellipse figure, also comprises:
Calculate the major axis of described ellipse figure and the angle of X-axis;
Determining that the behavior of described pig to be identified is for after couching, according to the horizontal ordinate of the left end point of described major axis, the horizontal ordinate of right endpoint, the direction that couches of pig to be identified described in the center point coordinate of described ellipse figure and described angle identification.
6. the method for identification pig lying behaviour according to claim 5, it is characterized in that, the horizontal ordinate of the described left end point according to described major axis, the horizontal ordinate of right endpoint, the direction that couches of pig to be identified described in the center point coordinate of described ellipse figure and described angle identification, comprising:
Pig to be identified with the standard deviation of all point of the described major axis Zhong Mei side, both sides that is separator bar to the distance of described major axis according to the center point coordinate of described ellipse figure and described angle calcu-lation;
Described standard deviation is compared, if be greater than back to all point of thermolamp side to the standard deviation of the distance of described major axis towards all point of thermolamp side to the standard deviation of the distance of described major axis, then determine that the direction that couches of described pig to be identified is that breast side couches towards thermolamp, otherwise, determine that the direction that couches of described pig to be identified is that dorsal part couches towards thermolamp.
7. identify a device for pig lying behaviour, it is characterized in that, comprising:
Image acquisition unit, for obtaining bianry image and the depth image of pig to be identified;
Coordinate calculating unit, for described bianry image is fitted to ellipse figure, and calculates the horizontal ordinate of the left end point of the major axis of described ellipse figure, the horizontal ordinate of right endpoint, the center point coordinate of described ellipse figure;
Height calculation unit, for calculate in described depth image cross described center point coordinate parallel with X-axis and on the line segment that the horizontal ordinate of left end point and right endpoint is equal with the horizontal ordinate of right endpoint with the left end point of described major axis respectively each pixel to the height on ground;
Recognition unit, for being compared to the maximal value in the height on ground and the numerical value preset by described pixel, if described maximal value is less than described default numerical value, determines that the behavior of described pig to be identified is for couching.
8. the device of identification pig lying behaviour according to claim 7, it is characterized in that, described image acquisition unit, comprising:
First Image Acquisition subelement, for obtaining the first digital picture under feeding environment and the second digital picture; Wherein, described first digital picture does not comprise pig to be identified, and described second digital picture comprises described pig to be identified;
Second Image Acquisition subelement, for by described second digital picture is deducted described first digital picture, obtain the digital picture of described pig to be identified, and the digital picture of described pig to be identified is carried out two-value conversion, obtain the bianry image of described pig to be identified.
9. the device of identification pig lying behaviour according to claim 7, is characterized in that, described coordinate calculating unit, for by minimum two-value matching, described bianry image is fitted to ellipse figure.
10. the device of identification pig lying behaviour according to claim 7, it is characterized in that, described height calculation unit, comprising:
First high computational subelement, for for each pixel on described line segment, this pixel calculated and straight line vertical with described line segment is the depth value of each pixel of 0,1 and 2 with the pixel distance of this pixel, and to calculate described according to formula h eight=h-d be that the pixel of 0,1 and 2 is to the height on ground with the pixel distance of this pixel, described height is the height of pixel to ground, described h is the height of camera to ground, and described d is the depth value of pixel;
Second high computational subelement, for for each pixel on described line segment, this pixel calculated and the straight line vertical with described line segment with the pixel distance of this pixel is the pixel of 0,1 and 2 to the average of the height on ground, and using this average as this pixel to the height on ground.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510232700.8A CN104881636B (en) | 2015-05-08 | 2015-05-08 | Identify the method and device of pig lying behaviour |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510232700.8A CN104881636B (en) | 2015-05-08 | 2015-05-08 | Identify the method and device of pig lying behaviour |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104881636A true CN104881636A (en) | 2015-09-02 |
| CN104881636B CN104881636B (en) | 2018-07-24 |
Family
ID=53949124
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510232700.8A Active CN104881636B (en) | 2015-05-08 | 2015-05-08 | Identify the method and device of pig lying behaviour |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104881636B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107844797A (en) * | 2017-09-27 | 2018-03-27 | 华南农业大学 | A kind of method of the milking sow posture automatic identification based on depth image |
| CN109164458A (en) * | 2018-08-28 | 2019-01-08 | 日立楼宇技术（广州）有限公司 | A kind of cultivation monitoring method and system |
| CN109766856A (en) * | 2019-01-16 | 2019-05-17 | 华南农业大学 | A kind of method of double fluid RGB-D Faster R-CNN identification milking sow posture |
| CN109964845A (en) * | 2019-04-26 | 2019-07-05 | 中国农业科学院农业信息研究所 | Piglet safety apparatus |
| CN110651728A (en) * | 2019-11-05 | 2020-01-07 | 秒针信息技术有限公司 | Piglet pressed detection method, device and system |
| CN111767794A (en) * | 2020-05-28 | 2020-10-13 | 华南农业大学 | Cage-rearing poultry abnormal behavior detection method and detection system based on machine vision |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050169555A1 (en) * | 2003-11-07 | 2005-08-04 | Yuichi Hasegawa | Image processing apparatus and method, and computer program |
| CN102509085A (en) * | 2011-11-19 | 2012-06-20 | 江苏大学 | Pig walking posture identification system and method based on outline invariant moment features |
| CN102521563A (en) * | 2011-11-19 | 2012-06-27 | 江苏大学 | Method for indentifying pig walking postures based on ellipse fitting |
| CN103824056A (en) * | 2014-02-18 | 2014-05-28 | 江苏大学 | Pig posture recognition method based on Zernike moment and support vector machine |
-
2015
- 2015-05-08 CN CN201510232700.8A patent/CN104881636B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050169555A1 (en) * | 2003-11-07 | 2005-08-04 | Yuichi Hasegawa | Image processing apparatus and method, and computer program |
| CN102509085A (en) * | 2011-11-19 | 2012-06-20 | 江苏大学 | Pig walking posture identification system and method based on outline invariant moment features |
| CN102521563A (en) * | 2011-11-19 | 2012-06-27 | 江苏大学 | Method for indentifying pig walking postures based on ellipse fitting |
| CN103824056A (en) * | 2014-02-18 | 2014-05-28 | 江苏大学 | Pig posture recognition method based on Zernike moment and support vector machine |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107844797A (en) * | 2017-09-27 | 2018-03-27 | 华南农业大学 | A kind of method of the milking sow posture automatic identification based on depth image |
| CN109164458A (en) * | 2018-08-28 | 2019-01-08 | 日立楼宇技术（广州）有限公司 | A kind of cultivation monitoring method and system |
| CN109766856A (en) * | 2019-01-16 | 2019-05-17 | 华南农业大学 | A kind of method of double fluid RGB-D Faster R-CNN identification milking sow posture |
| CN109766856B (en) * | 2019-01-16 | 2022-11-15 | 华南农业大学 | Method for recognizing postures of lactating sows through double-current RGB-D Faster R-CNN |
| CN109964845A (en) * | 2019-04-26 | 2019-07-05 | 中国农业科学院农业信息研究所 | Piglet safety apparatus |
| CN109964845B (en) * | 2019-04-26 | 2023-12-01 | 中国农业科学院农业信息研究所 | Piglet rescue device |
| CN110651728A (en) * | 2019-11-05 | 2020-01-07 | 秒针信息技术有限公司 | Piglet pressed detection method, device and system |
| CN110651728B (en) * | 2019-11-05 | 2022-08-09 | 秒针信息技术有限公司 | Piglet pressed detection method, device and system |
| CN111767794A (en) * | 2020-05-28 | 2020-10-13 | 华南农业大学 | Cage-rearing poultry abnormal behavior detection method and detection system based on machine vision |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104881636B (en) | 2018-07-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104881636A (en) | Method and device for identifying lying behavior of pig | |
| US20230281265A1 (en) | Method for estimating body size and weight of pig based on deep learning | |
| CN110598658B (en) | Convolutional network identification method for sow lactation behaviors | |
| CN103558910B (en) | A kind of intelligent display system of automatic tracking head pose | |
| CN113313833B (en) | Pig weight estimation method based on 3D vision technology | |
| CN112232978B (en) | Aquatic product length and weight detection method, terminal equipment and storage medium | |
| CN110490161B (en) | Captive animal behavior analysis method based on deep learning | |
| CN108470375A (en) | Nerve trachea automatic detection algorithm based on deep learning | |
| CN108182674A (en) | Granularity Detection analysis method based on U-Net deep learning networks | |
| CN105701809A (en) | Flat-field correction method based on line-scan digital camera scanning | |
| CN108765264A (en) | Image U.S. face method, apparatus, equipment and storage medium | |
| CN115100513B (en) | Method and system for estimating food intake of breeding object based on computer vision recognition | |
| CN108875644A (en) | A kind of aquatic animal density survey method | |
| CN108734054B (en) | Non-shielding citrus fruit image identification method | |
| CA3230401A1 (en) | Systems and methods for the automated monitoring of animal physiological conditions and for the prediction of animal phenotypes and health outcomes | |
| CN111833331A (en) | Livestock body temperature image processing method based on artificial intelligence and active gun-ball linkage | |
| CN113180640B (en) | Size measuring method and system for livestock breeding | |
| CN112906533B (en) | Safety helmet wearing detection method based on self-adaptive detection area | |
| CN108682000B (en) | Pig body length and body width detection method based on Kinect video | |
| CN112396016B (en) | Face recognition system based on big data technology | |
| CN113284111A (en) | Hair follicle region positioning method and system based on binocular stereo vision | |
| CN111369497B (en) | Walking type tree fruit continuous counting method and device | |
| CN110215001A (en) | One kind, which is cut the garment according to the figure, measures interactive accurate measurement method | |
| CN113449638B (en) | Pig image ideal frame screening method based on machine vision technology | |
| CN115619823A (en) | Object putting method, device, terminal and computer readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |