阅读说明：本技术 一种智能绵羊繁殖能力检测的系统及方法 (Intelligent sheep reproductive capacity detection system and method ) 是由 苗向阳 解领丽 刘天义 于 2020-05-07 设计创作，主要内容包括：本发明属于智能绵羊繁殖能力检测技术领域,公开了一种智能绵羊繁殖能力检测的系统及方法,包括：绵羊图像采集模块、体征信息采集模块、基因信息采集模块、主控模块、健康监控模块、繁殖数计算模块、繁殖评价模块、云服务模块、显示模块。本发明通过健康监控模块采集数据方法简单,数据内容丰富,数据挖掘后可产生可靠性较强的个体数据,有利于养殖场的人力精简、成本缩减和精确到个体的绵羊生长过程的行为监控和溯源,实现绵羊健康状态的实时监控；同时,通过繁殖数计算模块利用对拍摄图像进行灰度处理、拼接处理以及滤波、阈值处理等,将拍摄图像转换成二值图像,并在二值图像中进行轮廓检索以得到绵羊繁殖群体数目。(The invention belongs to the technical field of intelligent sheep reproductive capacity detection, and discloses an intelligent sheep reproductive capacity detection system and method, which comprises the following steps: the sheep breeding system comprises a sheep image acquisition module, a sign information acquisition module, a gene information acquisition module, a main control module, a health monitoring module, a breeding number calculation module, a breeding evaluation module, a cloud service module and a display module. The method for acquiring data through the health monitoring module is simple, the data content is rich, individual data with high reliability can be generated after data mining, the method is beneficial to simplifying manpower of a farm, reducing cost and accurately monitoring and tracing the behavior of an individual sheep growth process, and the real-time monitoring of the sheep health state is realized; meanwhile, the shot images are converted into binary images by a reproduction number calculation module through gray processing, splicing processing, filtering, threshold processing and the like of the shot images, and contour retrieval is carried out in the binary images to obtain the number of sheep reproduction groups.)

1. An intelligent sheep reproductive capacity detection method is characterized by comprising the following steps:

acquiring sheep breeding population image data by using a camera, and acquiring sheep breeding vital sign data by using a related image data processing program; collecting gene data of a breeding sheep by using a gene detection collector;

monitoring the health state of the sheep according to the acquired data;

step three, evaluating the fecundity of the sheep according to the health state of the sheep in the step two, and estimating the number of the sheep breeding;

step four, the data are transmitted to a cloud server, the collected sheep data are subjected to cloud processing by using related cloud computing, and meanwhile the cloud server transmits information to each terminal;

displaying the acquired sheep breeding population image, vital signs and gene data, and health, breeding number and evaluation result information by using a display;

in the second step, the sheep health collection method further comprises the following steps:

(1) the method comprises the following steps that each sheep is provided with an electronic tag, a corresponding electronic tag reader is arranged in a sheep activity area, and a camera device for acquiring sheep activity video images is arranged in the sheep activity area;

(2) acquiring the electronic tag identity of each sheep and individual movement track information corresponding to the electronic tag identity by using the electronic tag reader and the video image acquired by the camera device;

(3) acquiring the total displacement of each sheep at intervals according to the individual movement track information;

(4) acquiring the feeding condition of each sheep according to the position of a trough in the video image and the individual movement track information, wherein the feeding condition comprises the times of approaching the trough and the length of each feeding stay;

(5) and acquiring the drinking condition of each sheep according to the position of the water tank in the video image and the individual movement track information, wherein the drinking condition comprises the number of times of approaching the water tank and the length of stay of each drinking water.

2. The intelligent sheep fecundity detection method according to claim 1, wherein the step of acquiring the electronic tag identity of each sheep and the individual movement track information corresponding to the electronic tag identity by using the video images acquired by the electronic tag reader and the camera device specifically comprises the following steps:

acquiring the electronic tag identity of each sheep, the time when the electronic tag identity is identified and the geographic position of the electronic tag reader by using the electronic tag reader;

acquiring the visual movement track information of each sheep in the video image by using the video image;

traversing all the visual movement track information according to the time when the electronic tag identity of each sheep is identified and the geographic position of the electronic tag reader, acquiring the visual movement track information with the same time when the corresponding electronic tag identity is identified and the geographic position of the electronic tag reader, and matching the visual movement track information with the corresponding electronic tag identity to form individual movement track information corresponding to the electronic tag identity.

3. The intelligent sheep fecundity detection method according to claim 2, wherein the step of obtaining the visual movement track information of each sheep in the video image by using the video image specifically comprises the steps of:

identifying all sheep individuals in each frame of the video image by utilizing a deep learning technology; giving a visual number to each livestock in each frame, and if the current frame has the same sheep individual as that in the previous frame, giving the sheep individual in the current frame the same visual number as that in the previous frame;

and acquiring the frame position of each visual number in each frame of the video image and the time corresponding to each frame of the video image to form visual movement track information corresponding to the visual number, the frame position of the visual number in each frame of the video image and the time of each frame of the video image in a one-to-one manner.

4. The intelligent sheep fecundity detection method according to claim 1, wherein in the third step, the reproductive count calculation method is as follows:

1) acquiring a sheep breeding population shooting image through a camera;

2) preprocessing the shot image, wherein the preprocessing comprises parallel gray processing, splicing processing and Gaussian filtering processing;

3) carrying out threshold processing and morphological filtering processing on the preprocessed shot image to obtain a binary image of the sheep breeding population;

4) and carrying out contour retrieval on the binary image to obtain the number of sheep breeding groups.

5. The system for intelligent sheep fecundity detection of claim 4, wherein the parallel gray scale processing comprises:

the method comprises the steps of partitioning a shot image into blocks according to lines, establishing a corresponding relation between a brightness value Y and RGB three color components according to a change relation between RGB and YUV color space, namely Y is 0.3R +0.59G +0.11B, and obtaining the brightness value of each pixel in each block;

and taking the brightness value as a gray value of the shot image to obtain a gray image.

6. The intelligent sheep fecundity detection method of claim 4, wherein the splicing process comprises:

acquiring a frequency domain information image of the gray level image according to a parallel FFT algorithm;

acquiring a cross power spectrum of adjacent shot images according to the frequency domain information image of the gray level image;

reversely splicing the shot images according to the cross-power spectrum peak values of the adjacent shot images;

and optimizing the splicing result by using a gradual-out and gradual-in image fusion method.

7. The intelligent sheep fecundity detection method of claim 6, wherein the parallel FFT algorithm comprises:

dividing the gray level image into a plurality of blocks according to rows, and simultaneously carrying out parallel operation on each block according to a one-dimensional FFT processing algorithm according to the rows to obtain an intermediate image;

and dividing the intermediate image into a plurality of blocks according to columns, and simultaneously performing parallel operation on each block according to a one-dimensional FFT processing algorithm according to the columns to obtain a frequency domain information image of the gray level image.

8. The intelligent sheep fecundity detection method of claim 4, wherein the thresholding comprises:

comparing the gray value of each pixel point of the shot image with a preset threshold value;

when the gray value of a pixel point is smaller than or equal to a preset threshold, giving the gray value of the pixel point to a first gray value;

and when the gray value of the pixel point is greater than the preset threshold value, giving the gray value of the pixel point to a second gray value.

9. The intelligent sheep fecundity detection method according to claim 4, wherein the contour searching the binary image to obtain the sheep population number comprises:

judging whether the gray values of all adjacent pixel points of the pixel points with the gray values being the first gray value in the binary image are the first gray value or not;

when the gray values of all adjacent pixel points of the pixel points with the gray values being the first gray value in the binary image are not the first gray value, marking the pixel points as contour points;

taking the contour point as an initial contour point, and carrying out contour retrieval on all adjacent pixel points of the initial contour point according to the anticlockwise direction until returning to the initial contour point to finish contour retrieval;

repeating the steps until all the contour retrieval is completed;

the number of contours was taken as the number of sheep breeding colonies.

10. A system for intelligent sheep fertility testing implementing the intelligent sheep fertility testing method of claims 1-7, the system comprising:

the sheep breeding system comprises a sheep image acquisition module, a sign information acquisition module, a gene information acquisition module, a main control module, a health monitoring module, a breeding number calculation module, a breeding evaluation module, a cloud service module and a display module;

the sheep image acquisition module is connected with the main control module, acquires sheep breeding population image data through the camera, performs denoising enhancement on the acquired image data, moves the gray value template in the sheep population image to be processed, and overlaps the gray value template with the special position pixel in the image; determining the gray value of the pixel corresponding to the special position according to the contrast data of the gray value template and the pixel at the special position; finding out the middle data in the gray values by using a related program; correcting each pixel point on the image by taking the intermediate data as a standard value;

the physical sign information acquisition module is connected with the main module and is used for extracting data capable of reflecting physical signs of the sheep from the acquired sheep image data by using a related program; judging sheep sign information according to the image data of the sheep signs;

the gene information acquisition module is connected with the main control module, acquires gene data of the breeding sheep by using the gene detection collector, and extracts the sheep physique characteristic information according to the acquired gene information;

the main control module is connected with the sheep image acquisition module, the physical sign information acquisition module, the gene information acquisition module, the health monitoring module, the breeding number calculation module, the breeding evaluation module, the cloud service module and the display module and is used for controlling the modules to normally work through a host;

the health monitoring module is connected with the main module, a corresponding electronic tag reader is arranged in the sheep activity area, and a camera device for acquiring sheep activity video images is arranged in the sheep activity area; acquiring the electronic tag identity of each sheep and individual movement track information corresponding to the electronic tag identity by using the electronic tag reader and the video image acquired by the camera device; acquiring the feeding condition of each sheep according to the position of a trough in the video image and the individual movement track information, wherein the feeding condition comprises the times of approaching the trough and the length of each feeding stay; acquiring the drinking condition of each sheep according to the position of a water tank in the video image and the individual movement track information, wherein the drinking condition comprises the number of times of approaching the water tank and the length of stay of each drinking, and the health of breeding sheep is monitored;

the breeding number calculation module is connected with the main module, acquires a sheep breeding population shooting image through a camera, performs preprocessing on the shooting image, performs threshold processing and morphological filtering processing on the preprocessed shooting image to acquire a binary image of the sheep breeding population, and performs contour retrieval on the binary image to acquire the number of the sheep breeding population;

the reproduction evaluation module is connected with the main module and used for evaluating the reproduction capacity of the sheep according to the collected data through an evaluation program;

the cloud server receives the data, performs cloud processing on the acquired sheep data by using related cloud computing, and simultaneously transmits information to each terminal;

and the display module is connected with the main module and used for displaying the collected sheep breeding population images, vital signs, gene data, health, breeding number and evaluation result information through a display.

Technical Field

The invention belongs to the technical field of intelligent sheep reproductive capacity detection, and particularly relates to an intelligent sheep reproductive capacity detection system and method.

Background

The sheep body is full, the hair is thick and the head is short. The male sheep mostly has spiral large angles and is deterrent, and the female sheep has no angles or small angles. The skull has a lacrimal fossa, and the nasal bone is relatively raised. All four hoofs have toe glands. The male sheep has no qi. The weight of the body varies from ten kilograms to more than one hundred kilograms. The structure and habit of the utility model are suitable for grazing due to various characteristics; the mouth tip and the lips are thin and flexible, which is beneficial to eating short grass and can also eat thick and hard straws and branches; the digestion capability is strong; some species can accumulate fat around the tail, buttocks and internal organs for consumption in the absence of green feed in winter and spring; the imitative and the uniting property are strong, and the habit of following the uniting and the uniting of the collarband sheep (usually the old female sheep) exists; when grazing, the user can take food from high place and is happy to sleep at high place at night. The quilt can resist cold and heat due to the heat preservation and insulation function of the quilt; but soon after shearing, it is vulnerable to disease if the weather is cold or is caught in rain. Generally, people like dry but afraid of damp-heat. Weak property, weak self-defense ability and easy damage by animals. The natural life is about 15 years old. However, the health of the sheep cannot be accurately monitored in the existing intelligent sheep reproductive capacity detection process; meanwhile, counting sheep breeding groups manually wastes time and labor, and calculation is inaccurate, so that the prediction of the sheep breeding capacity is influenced.

In summary, the problems of the prior art are as follows:

the existing intelligent sheep reproductive capacity detection process cannot accurately monitor the sheep health; meanwhile, counting sheep breeding groups manually wastes time and labor, and calculation is inaccurate, so that the prediction of the sheep breeding capacity is influenced.

Disclosure of Invention

Aiming at the problems in the prior art, the invention provides an intelligent sheep reproductive capacity detection system and method.

The invention is realized in such a way that an intelligent sheep reproductive capacity detection method comprises the following steps:

acquiring sheep breeding population image data by using a camera, and acquiring sheep breeding vital sign data by using a related image data processing program; collecting gene data of a breeding sheep by using a gene detection collector;

monitoring the health state of the sheep according to the acquired data;

step three, evaluating the fecundity of the sheep according to the health state of the sheep in the step two, and estimating the number of the sheep breeding;

step four, the data are transmitted to a cloud server, the collected sheep data are subjected to cloud processing by using related cloud computing, and meanwhile the cloud server transmits information to each terminal;

and fifthly, displaying the acquired sheep breeding population image, the vital signs and the gene data as well as health, breeding number and evaluation result information by using a display.

In the second step, the sheep health assessment method further comprises the following steps:

(1) the method comprises the following steps that each sheep is provided with an electronic tag, a corresponding electronic tag reader is arranged in a sheep activity area, and a camera device for acquiring sheep activity video images is arranged in the sheep activity area;

(2) acquiring the electronic tag identity of each sheep and individual movement track information corresponding to the electronic tag identity by using the electronic tag reader and the video image acquired by the camera device;

(3) acquiring the total displacement of each sheep at intervals according to the individual movement track information;

(4) acquiring the feeding condition of each sheep according to the position of a trough in the video image and the individual movement track information, wherein the feeding condition comprises the times of approaching the trough and the length of each feeding stay;

(5) and acquiring the drinking condition of each sheep according to the position of the water tank in the video image and the individual movement track information, wherein the drinking condition comprises the number of times of approaching the water tank and the length of stay of each drinking water.

Further, the step of acquiring the electronic tag identity of each sheep and the individual movement track information corresponding to the electronic tag identity by using the video image acquired by the electronic tag reader and the camera device specifically includes the following steps:

acquiring the electronic tag identity of each sheep, the time when the electronic tag identity is identified and the geographic position of the electronic tag reader by using the electronic tag reader;

acquiring the visual movement track information of each sheep in the video image by using the video image;

traversing all the visual movement track information according to the time when the electronic tag identity of each sheep is identified and the geographic position of the electronic tag reader, acquiring the visual movement track information with the same time when the corresponding electronic tag identity is identified and the geographic position of the electronic tag reader, and matching the visual movement track information with the corresponding electronic tag identity to form individual movement track information corresponding to the electronic tag identity.

Further, the acquiring of the visual movement track information of each sheep in the video image by using the video image specifically includes the following steps:

identifying all sheep individuals in each frame of the video image by utilizing a deep learning technology; giving a visual number to each livestock in each frame, and if the current frame has the same sheep individual as that in the previous frame, giving the sheep individual in the current frame the same visual number as that in the previous frame;

and acquiring the frame position of each visual number in each frame of the video image and the time corresponding to each frame of the video image to form visual movement track information corresponding to the visual number, the frame position of the visual number in each frame of the video image and the time of each frame of the video image in a one-to-one manner.

Further, in the third step, the propagation number calculation method is as follows:

1) acquiring a sheep breeding population shooting image through a camera;

2) preprocessing the shot image, wherein the preprocessing comprises parallel gray processing, splicing processing and Gaussian filtering processing;

3) carrying out threshold processing and morphological filtering processing on the preprocessed shot image to obtain a binary image of the sheep breeding population;

4) and carrying out contour retrieval on the binary image to obtain the number of sheep breeding groups.

5. The system for intelligent sheep fecundity detection of claim 4, wherein the parallel gray scale processing comprises:

the method comprises the steps of partitioning a shot image into blocks according to lines, establishing a corresponding relation between a brightness value Y and RGB three color components according to a change relation between RGB and YUV color space, namely Y is 0.3R +0.59G +0.11B, and obtaining the brightness value of each pixel in each block;

and taking the brightness value as a gray value of the shot image to obtain a gray image.

Further, the splicing process includes:

acquiring a frequency domain information image of the gray level image according to a parallel FFT algorithm;

acquiring a cross power spectrum of adjacent shot images according to the frequency domain information image of the gray level image;

reversely splicing the shot images according to the cross-power spectrum peak values of the adjacent shot images;

and optimizing the splicing result by using a gradual-out and gradual-in image fusion method.

Further, the parallel FFT algorithm includes:

dividing the gray level image into a plurality of blocks according to rows, and simultaneously carrying out parallel operation on each block according to a one-dimensional FFT processing algorithm according to the rows to obtain an intermediate image;

and dividing the intermediate image into a plurality of blocks according to columns, and simultaneously performing parallel operation on each block according to a one-dimensional FFT processing algorithm according to the columns to obtain a frequency domain information image of the gray level image.

Further, the thresholding comprises:

comparing the gray value of each pixel point of the shot image with a preset threshold value;

when the gray value of a pixel point is smaller than or equal to a preset threshold, giving the gray value of the pixel point to a first gray value;

and when the gray value of the pixel point is greater than the preset threshold value, giving the gray value of the pixel point to a second gray value.

Further, the performing contour retrieval on the binary image to obtain the number of sheep groups comprises:

judging whether the gray values of all adjacent pixel points of the pixel points with the gray values being the first gray value in the binary image are the first gray value or not;

when the gray values of all adjacent pixel points of the pixel points with the gray values being the first gray value in the binary image are not the first gray value, marking the pixel points as contour points;

taking the contour point as an initial contour point, and carrying out contour retrieval on all adjacent pixel points of the initial contour point according to the anticlockwise direction until returning to the initial contour point to finish contour retrieval;

repeating the steps until all the contour retrieval is completed;

the number of contours was taken as the number of sheep breeding colonies.

The invention also provides a system for intelligent sheep fecundity detection, which implements the intelligent sheep fecundity detection method, and the system for intelligent sheep fecundity detection comprises:

the sheep breeding system comprises a sheep image acquisition module, a sign information acquisition module, a gene information acquisition module, a main control module, a health monitoring module, a breeding number calculation module, a breeding evaluation module, a cloud service module and a display module;

the sheep image acquisition module is connected with the main control module, acquires sheep breeding population image data through the camera, performs denoising enhancement on the acquired image data, moves the gray value template in the sheep population image to be processed, and overlaps the gray value template with the special position pixel in the image; determining the gray value of the pixel corresponding to the special position according to the contrast data of the gray value template and the pixel at the special position; finding out the middle data in the gray values by using a related program; correcting each pixel point on the image by taking the intermediate data as a standard value;

the physical sign information acquisition module is connected with the main module and is used for extracting data capable of reflecting physical signs of the sheep from the acquired sheep image data by using a related program; judging sheep sign information according to the image data of the sheep signs;

the gene information acquisition module is connected with the main control module, acquires gene data of the breeding sheep by using the gene detection collector, and extracts the sheep physique characteristic information according to the acquired gene information;

the main control module is connected with the sheep image acquisition module, the physical sign information acquisition module, the gene information acquisition module, the health monitoring module, the breeding number calculation module, the breeding evaluation module, the cloud service module and the display module and is used for controlling the modules to normally work through a host;

the health monitoring module is connected with the main module, a corresponding electronic tag reader is arranged in the sheep activity area, and a camera device for acquiring sheep activity video images is arranged in the sheep activity area; acquiring the electronic tag identity of each sheep and individual movement track information corresponding to the electronic tag identity by using the electronic tag reader and the video image acquired by the camera device; acquiring the feeding condition of each sheep according to the position of a trough in the video image and the individual movement track information, wherein the feeding condition comprises the times of approaching the trough and the length of each feeding stay; acquiring the drinking condition of each sheep according to the position of a water tank in the video image and the individual movement track information, wherein the drinking condition comprises the number of times of approaching the water tank and the length of stay of each drinking, and the health of breeding sheep is monitored;

the breeding number calculation module is connected with the main module, acquires a sheep breeding population shooting image through a camera, performs preprocessing on the shooting image, performs threshold processing and morphological filtering processing on the preprocessed shooting image to acquire a binary image of the sheep breeding population, and performs contour retrieval on the binary image to acquire the number of the sheep breeding population;

the reproduction evaluation module is connected with the main module and used for evaluating the reproduction capacity of the sheep according to the collected data through an evaluation program;

the cloud server receives the data, performs cloud processing on the acquired sheep data by using related cloud computing, and simultaneously transmits information to each terminal;

and the display module is connected with the main module and used for displaying the collected sheep breeding population images, vital signs, gene data, health, breeding number and evaluation result information through a display.

The invention has the advantages and positive effects that:

the method for acquiring data through the health monitoring module is simple, the data content is rich, individual data with high reliability can be generated after data mining, the method is beneficial to simplifying manpower of a farm, reducing cost and accurately monitoring and tracing the behavior of an individual sheep growth process, scientific and individualized sheep breeding management is realized, and real-time monitoring of sheep health states is realized; meanwhile, the shot images are converted into binary images by a reproduction number calculation module through gray processing, splicing processing, filtering, threshold processing and the like of the shot images, and contour retrieval is carried out in the binary images to obtain the number of sheep reproduction groups.

In the preprocessing process of acquiring the sheep breeding colony image data, the noise elimination method is adopted, so that the most real data can be obtained at the edge of the image, and the most accurate data of the image is reserved.

Drawings

FIG. 1 is a flow chart of a method for intelligent sheep fertility detection provided by the embodiment of the invention.

Fig. 2 is a block diagram of a system structure for detecting the reproductive capacity of an intelligent sheep according to an embodiment of the invention.

In fig. 2: 1. a sheep image acquisition module; 2. a physical sign information acquisition module; 3. a gene information acquisition module; 4. a main control module; 5. a health monitoring module; 6. a propagation number calculation module; 7. a propagation evaluation module; 8. a cloud service module; 9. and a display module.

Detailed Description

In order to further understand the contents, features and effects of the present invention, the following embodiments are illustrated and described in detail with reference to the accompanying drawings.

The structure of the present invention will be described in detail below with reference to the accompanying drawings.

As shown in figure 1, the intelligent sheep fecundity detection method provided by the invention comprises the following steps:

s101: the method comprises the steps that image data of a sheep breeding population are collected through a camera, and vital sign data of breeding sheep are collected through a related image data processing program; the gene data of the breeding sheep are collected by utilizing a gene detection collector.

S102: and monitoring the health state of the sheep according to the acquired data.

S103: and (5) evaluating the fecundity of the sheep according to the health state of the sheep in the step two, and estimating the number of the sheep reproduction.

S104: and transmitting the data to a cloud server, carrying out cloud processing on the acquired sheep data by using related cloud computing, and transmitting the information to each terminal by the cloud server.

S105: the display is used for displaying the collected sheep breeding colony images, vital signs and gene data as well as health, breeding number and evaluation result information.

As shown in fig. 2, the system for detecting the reproductive capacity of an intelligent sheep provided by the embodiment of the present invention includes: the sheep breeding system comprises a sheep image acquisition module, a sign information acquisition module, a gene information acquisition module, a main control module, a health monitoring module, a breeding number calculation module, a breeding evaluation module, a cloud service module and a display module;

the sheep image acquisition module is connected with the main control module, acquires sheep breeding population image data through the camera, performs denoising enhancement on the acquired image data, moves the gray value template in the sheep population image to be processed, and overlaps the gray value template with the special position pixel in the image; determining the gray value of the pixel corresponding to the special position according to the contrast data of the gray value template and the pixel at the special position; finding out the middle data in the gray values by using a related program; and taking the intermediate data as a standard value, and correcting each pixel point on the image.

The physical sign information acquisition module is connected with the main module and is used for extracting data capable of reflecting physical signs of the sheep from the acquired sheep image data by using a related program; and judging the sheep sign information according to the image data of the sheep signs.

And the gene information acquisition module is connected with the main control module, acquires gene data of the breeding sheep by using the gene detection collector, and extracts the sheep physique characteristic information according to the acquired gene information.

The main control module is connected with the sheep image acquisition module, the sign information acquisition module, the gene information acquisition module, the health monitoring module, the breeding number calculation module, the breeding evaluation module, the cloud service module and the display module and is used for controlling the modules to normally work through a host.

The health monitoring module is connected with the main module, a corresponding electronic tag reader is arranged in the sheep activity area, and a camera device for acquiring sheep activity video images is arranged in the sheep activity area; acquiring the electronic tag identity of each sheep and individual movement track information corresponding to the electronic tag identity by using the electronic tag reader and the video image acquired by the camera device; acquiring the feeding condition of each sheep according to the position of a trough in the video image and the individual movement track information, wherein the feeding condition comprises the times of approaching the trough and the length of each feeding stay; and acquiring the drinking condition of each sheep according to the position of the water tank in the video image and the individual movement track information, wherein the drinking condition comprises the number of times of approaching the water tank and the length of stay of each drinking, so as to realize monitoring on the health of breeding sheep.

The breeding number calculation module is connected with the main module, acquires a sheep breeding colony shooting image through a camera, performs preprocessing on the shooting image, wherein the preprocessing comprises parallel gray processing, splicing processing and Gaussian filtering processing, performs threshold processing and morphological filtering processing on the preprocessed shooting image to acquire a binary image of the sheep breeding colony, and performs contour retrieval on the binary image to acquire the number of the sheep breeding colony.

And the reproduction evaluation module is connected with the main module and used for evaluating the reproduction capacity of the sheep according to the collected data through an evaluation program.

And the cloud server receives the data, performs cloud processing on the acquired sheep data by using related cloud computing, and transmits the information to each terminal.

And the display module is connected with the main module and used for displaying the collected sheep breeding population images, vital signs, gene data, health, breeding number and evaluation result information through a display.

The sheep breeding population image processing method is characterized in that the sheep breeding population image processing method comprises the following specific processes of processing sheep breeding population images by the sheep image acquisition module 1 which is connected with the main module 4 and is used for acquiring sheep breeding population image data through a camera:

moving the gray value template in the sheep colony image to be processed, and overlapping the gray value template with the special position pixel in the image; determining the gray value of the pixel corresponding to the special position according to the contrast data of the gray value template and the pixel at the special position;

arranging the gray value data measured by the sheep group images into a line from small to large according to a certain sequence;

finding out the middle data in the gray values by using a related program; and taking the intermediate data as a standard value, and correcting each pixel point on the image to obtain an image enhanced image.

The sheep image noise eliminating method is connected with a sheep image acquisition module 1, a physical sign information acquisition module 2, a gene information acquisition module 3, a health monitoring module 5, a reproduction number calculation module 6, a reproduction evaluation module 7, a cloud service module 8 and a display module 9, and is used for controlling the main control module 4 of each module to normally work through a host to eliminate noise in sheep colony images in the following specific process:

each module maximum value on a certain scale L is an assumed signal module maximum value point, and L +1 is set as a search interval in a certain range according to set standard parameters; taking a point on a certain scale L as an origin point, and taking a range of a certain scale as a search;

detecting points in each search interval, and judging whether a module maximum value point exists or not; the maximum value point on the scale L is the maximum value point of the model formed by the noise;

finding a maximum value point in a search interval, and taking the maximum value point as a mode maximum value point formed by noise;

and overlapping the modulus maximum point in the search interval with the origin on the scale L to obtain a real edge.

The invention provides a method for reconstructing a sheep image without noise, which specifically comprises the following steps:

respectively transforming the sheep images without the noise by using wavelets to establish wavelet tower type decomposition of the images;

performing fusion processing on each decomposition layer by adopting different fusion operators respectively to obtain a fused wavelet pyramid;

and performing wavelet reconstruction on the wavelet pyramid obtained after fusion to finally obtain an enhanced reconstructed image.

The monitoring method of the health monitoring module 5 provided by the invention comprises the following steps:

(1) the method comprises the following steps that each sheep is provided with an electronic tag, a corresponding electronic tag reader is arranged in a sheep activity area, and a camera device for acquiring sheep activity video images is arranged in the sheep activity area;

(2) acquiring the electronic tag identity of each sheep and individual movement track information corresponding to the electronic tag identity by using the electronic tag reader and the video image acquired by the camera device;

(3) acquiring the total displacement of each sheep at intervals according to the individual movement track information;

(4) acquiring the feeding condition of each sheep according to the position of a trough in the video image and the individual movement track information, wherein the feeding condition comprises the times of approaching the trough and the length of each feeding stay;

(5) and acquiring the drinking condition of each sheep according to the position of the water tank in the video image and the individual movement track information, wherein the drinking condition comprises the number of times of approaching the water tank and the length of stay of each drinking water.

The method for acquiring the electronic tag identity of each sheep and the individual movement track information corresponding to the electronic tag identity by using the electronic tag reader and the video image acquired by the camera device specifically comprises the following steps:

acquiring the electronic tag identity of each sheep, the time when the electronic tag identity is identified and the geographic position of the electronic tag reader by using the electronic tag reader;

acquiring the visual movement track information of each sheep in the video image by using the video image;

traversing all the visual movement track information according to the time when the electronic tag identity of each sheep is identified and the geographic position of the electronic tag reader, acquiring the visual movement track information with the same time when the corresponding electronic tag identity is identified and the geographic position of the electronic tag reader, and matching the visual movement track information with the corresponding electronic tag identity to form individual movement track information corresponding to the electronic tag identity.

The method for acquiring the visual movement track information of each sheep in the video image by using the video image specifically comprises the following steps:

identifying all sheep individuals in each frame of the video image by utilizing a deep learning technology; giving a visual number to each livestock in each frame, and if the current frame has the same sheep individual as that in the previous frame, giving the sheep individual in the current frame the same visual number as that in the previous frame;

and acquiring the frame position of each visual number in each frame of the video image and the time corresponding to each frame of the video image to form visual movement track information corresponding to the visual number, the frame position of the visual number in each frame of the video image and the time of each frame of the video image in a one-to-one manner.

The calculation method of the propagation number calculation module 6 provided by the invention is as follows:

1) acquiring a sheep breeding population shooting image through a camera;

2) preprocessing the shot image, wherein the preprocessing comprises parallel gray processing, splicing processing and Gaussian filtering processing;

3) carrying out threshold processing and morphological filtering processing on the preprocessed shot image to obtain a binary image of the sheep breeding population;

4) and carrying out contour retrieval on the binary image to obtain the number of sheep breeding groups.

The parallel gray scale processing provided by the invention comprises the following steps:

the method comprises the steps of partitioning a shot image into blocks according to lines, establishing a corresponding relation between a brightness value Y and RGB three color components according to a change relation between RGB and YUV color space, namely Y is 0.3R +0.59G +0.11B, and obtaining the brightness value of each pixel in each block;

and taking the brightness value as a gray value of the shot image to obtain a gray image.

The splicing treatment provided by the invention comprises the following steps:

acquiring a frequency domain information image of the gray level image according to a parallel FFT algorithm;

acquiring a cross power spectrum of adjacent shot images according to the frequency domain information image of the gray level image;

reversely splicing the shot images according to the cross-power spectrum peak values of the adjacent shot images;

and optimizing the splicing result by using a gradual-out and gradual-in image fusion method.

The parallel FFT algorithm provided by the invention comprises the following steps:

dividing the gray level image into a plurality of blocks according to rows, and simultaneously carrying out parallel operation on each block according to a one-dimensional FFT processing algorithm according to the rows to obtain an intermediate image;

and dividing the intermediate image into a plurality of blocks according to columns, and simultaneously performing parallel operation on each block according to a one-dimensional FFT processing algorithm according to the columns to obtain a frequency domain information image of the gray level image.

The threshold processing provided by the invention comprises the following steps:

comparing the gray value of each pixel point of the shot image with a preset threshold value;

when the gray value of a pixel point is smaller than or equal to a preset threshold, giving the gray value of the pixel point to a first gray value;

and when the gray value of the pixel point is greater than the preset threshold value, giving the gray value of the pixel point to a second gray value.

The invention provides a method for carrying out contour retrieval on the binary image to obtain the number of sheep groups, which comprises the following steps:

judging whether the gray values of all adjacent pixel points of the pixel points with the gray values being the first gray value in the binary image are the first gray value or not;

when the gray values of all adjacent pixel points of the pixel points with the gray values being the first gray value in the binary image are not the first gray value, marking the pixel points as contour points;

taking the contour point as an initial contour point, and carrying out contour retrieval on all adjacent pixel points of the initial contour point according to the anticlockwise direction until returning to the initial contour point to finish contour retrieval;

repeating the steps until all the contour retrieval is completed;

the number of contours was taken as the number of sheep breeding colonies.

The above description is only for the preferred embodiment of the present invention, and is not intended to limit the present invention in any way, and all simple modifications, equivalent changes and modifications made to the above embodiment according to the technical spirit of the present invention are within the scope of the technical solution of the present invention.