A livestock statistics monitoring system
Through the combination of photoelectric switches and image recognition technology, the inaccurate counting problem caused by the tailing of livestock through cards is solved, and accurate monitoring of the number of sows and the status of estrus is achieved, reducing manpower investment and economic losses.
Patent Information
- Application Number
- CN202211601967.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-12-13
- Publication Date
- 2024-12-20
- Estimated Expiration
- 2042-12-13
AI Technical Summary
In the animal husbandry industry, the inaccurate counting caused by livestock following the card is inaccurate, especially in the statistics of sow counts, which are prone to misidentification or misidentification, resulting in management burden and economic losses.
The combination of photoelectric switches and image acquisition module is adopted to identify the profile and behavioral characteristics of livestock through the counting channel and data processing module, generate occlusion signals and estrus indicators, correct the counting and screen the estrus sows to reduce manual observation errors.
The accurate count of livestock numbers and accurate identification of sow estrus status are achieved, which reduces manpower and material investment, improves statistics accuracy and efficiency, and reduces economic losses.
Smart Images
Figure CN115918569B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the field of animal husbandry, and in particular to an animal husbandry statistical monitoring system. Background Art
[0002] Livestock is the animals that have been domesticated by humans through artificial breeding, such as cattle, sheep, pigs, etc., through breeding to obtain livestock products such as meat, eggs, milk, wool, cashmere, hides, silk and medicinal materials. Livestock is an important part of agriculture, and is listed as the two pillars of agricultural production together with planting. Among livestock meat products, China's livestock breed structure is basically the same as the world's livestock breed structure. In order of proportion from high to low, they are pork, poultry, beef, and mutton. However, pork production accounts for 65% of China's total meat production, which is much higher than the proportion of other livestock species. In large-scale animal husbandry, statistics on the huge number of breeding individuals is a difficult problem to solve. Inaccurate statistics will bring great management burden and economic losses. Taking pig farm breeding as an example, the success or failure of pig farm operation depends on the average number of piglets provided by each sow per year and the number of commercial pigs contributed per year. Normal estrus and timely breeding of sows are the key to the success or failure of sow breeding. The sow breeding capacity of a pig farm with 10,000 heads is about 500. With such a large number of sows, it is difficult for breeders to identify the estrus status of each pig, and misidentification or missed identification may cause huge losses.
[0003] The invention patent with the patent number of 201810049114.3 discloses a livestock statistics monitoring system, including a bottom plate, the top of the bottom plate is fixedly connected to a breeding shed body, the top of the bottom plate and one side of the breeding shed body is fixedly connected to a U-shaped pass frame, the two sides of the inner wall of the U-shaped pass frame are rotatably connected to a rotating shaft through a bearing, and the outer surfaces of both ends of the rotating shaft are sleeved with worm sleeves, and the outer surface of the rotating shaft located between the two worm sleeves is sleeved with a first bevel gear, which relates to the field of livestock statistics monitoring technology. The livestock statistics monitoring system greatly facilitates the breeding staff to count the number of cattle and sheep returning to the shed, and well avoids the occurrence of statistical errors. The breeding staff does not need to spend a lot of time to count the cattle and sheep returning to the shed, which reduces the workload of the breeding staff, improves the work efficiency of the breeding staff, and realizes the automatic counting of cattle and sheep returning to the shed both quickly and accurately.
[0004] The above technology can complete the counting work when the livestock enter the breeding shed in an orderly manner. However, if the livestock are not orderly or have bad tempers, there may be more than one livestock passing through the U-shaped pass at the same time, or the later livestock may follow the previous livestock, resulting in the above technology The device for automatic counting through photoelectric sensing can only collect one occlusion signal, resulting in counting errors. In order to reduce the error of manual observation, a technology is now needed that can accurately count the number of sows and screen out estrus sows. Summary of the invention
[0005] The invention provides a livestock statistics monitoring system, which can solve the problem of inaccurate counting caused by livestock following through checkpoints.
[0006] In order to solve the above technical problems, this application provides the following technical solutions:
[0007] A livestock statistics monitoring system includes a photoelectric switch, the photoelectric switch is used to count livestock passing through the photoelectric switch and generate a first counting signal, and also includes:
[0008] Counting module: comprising a counting channel, the counting channel is used to accommodate livestock passing through, the photoelectric switch is fixed inside the counting channel, the width of the counting channel is adjustable, and the maximum width of the counting channel does not exceed 2 times the width of the livestock body;
[0009] Image acquisition module: The image acquisition module is used to acquire image information / video information of livestock in the counting channel; the image information / video information at least includes information before the livestock enters the photoelectric switch and when the livestock enters the photoelectric switch;
[0010] Data processing module: The data processing module is used to obtain image information / video information, cluster the pixels in the image, and extract the contour of the livestock; the data processing module analyzes the contour of the livestock near the photoelectric switch, and before the livestock enters the photoelectric switch, if the pixels of the two livestock in front and behind intersect, and the ratio of the number of pixels at the intersection to the number of pixels of one of the livestock exceeds a threshold, a first blocking signal is generated; when the livestock enters the photoelectric switch, if the pixels of the two livestock in front and behind intersect, and the ratio of the number of pixels at the intersection to the number of pixels of one of the livestock exceeds a threshold, a second blocking signal is generated;
[0011] Among them, the data processing module also obtains the first counting signal of the photoelectric switch in real time. When the data processing module receives the first blocking signal, the current first counting signal is locked; if the data processing module receives the second blocking signal again, and the new value of the first counting signal does not match the actual number of livestock passing through, it is determined that the counting is missing, and the number of consecutive second blocking signals is accumulated to form a new first counting signal.
[0012] Basic principles and beneficial effects: Photoelectric switch (i.e. photoelectric sensor) is the abbreviation of photoelectric proximity switch. It uses the blocking or reflection of the light beam by the detected object to connect the circuit by the synchronous circuit to detect the presence or absence of the object. When a single livestock passes through the counting channel, its body will first block the light beam, and after passing through, the light beam returns to normal, thus completing a count. The width of the counting channel is not enough to allow two pigs to pass side by side at the same time. The image acquisition module obtains the image information in the channel. The data processing module identifies and analyzes the images of the livestock before entering the photoelectric switch and when the photoelectric switch is turned on. The contour information of the two livestock in front and behind that are heading towards the photoelectric switch (not yet entered) in the image information / video information is identified. If the two livestock have an intersection of pixels and the number of intersections exceeds the threshold, it is determined that there is occlusion. For the occlusion before entering the photoelectric switch and when entering the photoelectric switch, the actual number of passes in the image or video is compared with the number of counts of the photoelectric switch. If the photoelectric switch only records one data due to occlusion, and the second occlusion signal indicates that only one signal is recorded when two or more livestock pass through the photoelectric switch due to occlusion, resulting in inaccurate counting, the number of second occlusion signals is superimposed on the original first counting signal to complete the counting correction. If only image recognition is used to complete counting, the computational requirements of the data processing module will increase, which is not conducive to popularization. The method of counting mainly with photoelectric switches and supplemented by image inspection and recognition can improve the accuracy of counting, and the cost investment is less than that of the full image recognition method. If only image recognition is relied upon, then space occlusion and animal movement will lead to inaccurate recognition. This solution only recognizes the designated area and judges congestion, which is more accurate and reliable.
[0013] This solution uses a combination of photoelectric switches and image acquisition modules to correct the number of livestock, thereby completing accurate counting and reducing the occurrence of errors. It also reduces the time, physical strength, and energy costs of human clearing. The use of data processing modules for intelligent control is conducive to scale expansion and management. In summary, the problem of inaccurate counting caused by livestock following the checkpoint is solved.
[0014] Furthermore, the livestock is a sow, and the data processing module is further used to extract contour information of the sow's ear based on the image information / video information. If the contour of the sow's ear presents an upwardly protruding sharp corner, a first estrus indicator is generated.
[0015] Beneficial effects: The success or failure of pig farm operation depends on the average number of piglets and commercial pigs provided by each sow per year. Normal estrus and timely mating of sows are the key to the success of sow farming. The sow breeding capacity of a 10,000-head pig farm is about 500. Under such a large number, it is difficult for farmers to identify the estrus status of each pig, and misidentification or missed identification may cause great losses. In the process of sow transportation, the estrus status of sows can be identified at the same time, which can save a lot of time and energy for breeders. When the state of the sow's ears is erected through image recognition, sows with signs of estrus can be preliminarily screened out.
[0016] Furthermore, a boar odorant is provided in the counting channel where the sow moves toward the photoelectric switch, and the boar odorant is used to attract the sow. Press sensors are provided on both sides of the channel, and the press sensors include a press switch, a power supply, a protective resistor, and an ammeter. The power supply, the press switch, the protective resistor and the ammeter are electrically connected in sequence.
[0017] Beneficial effects: Boar odor is a sex pheromone for pigs. When sows in estrus receive the smell of a boar, they will react and move toward the source of the smell. By attracting the sows with the smell of the boar, sows with signs of estrus can be further screened out. Press sensors are set on both sides of the channel. When a sow passes through the channel alone, she cannot squeeze the two press sensors on both sides at the same time due to her body shape. If both opposing press sensors are triggered, it can be determined that there is squeezing between the sows, which can assist in image recognition. Boar odor is applied to the surface of the press sensor. When the unilateral press sensor is triggered, the position in the image that contacts the press sensor is obtained for judgment. If it is the head of a pig, it can be determined that the sow is attracted by the boar odor and the press sensor is triggered by a sow in estrus. Compared with image information judgment, it is more accurate.
[0018] Furthermore, after the sow is attracted by the boar odorant, the image acquisition module is also used to acquire the state characteristics of the sow, and the state characteristics include the changes in the sow's body shape before and after being attracted. If the sow remains still for more than a threshold time after acquiring the boar odorant, a second estrus indicator is generated; the image acquisition module is also used to acquire the contour information of the sow's tail, and the data processing module constructs a line model with the same trend based on the contour information of the sow's tail, and establishes a coordinate system at the last turning point of the line model toward the tip of the sow's tail. If the tip of the tail is facing upward, a third estrus indicator is generated.
[0019] Beneficial effect: Another indicator of sow estrus is that the tail of a sow in estrus will curl upward. In the process of sows in the counting channel, the heads of the subsequent sows are likely to cover the tail status of the previous sow, that is, the image acquisition module cannot capture the tail status of the previous sow. Through the establishment of the boar attractant, if the previous sow shows signs of estrus, it will move toward the source of the smell, that is, it will deviate from the original movement trajectory, thus turning sideways and separating from the sow behind. The tail status is easier to be captured by the image acquisition module, and the sow will feel excited after smelling the boar odorant, and may accelerate to walk towards the source of the smell, further separating from the sow behind, so that the tail is completely exposed and captured by the image acquisition module. Further judgment of the estrus status of the sow is convenient for the breeding staff to deal with it.
[0020] Furthermore, facial feature information of each sow is extracted in advance to establish a database, and the data processing module assigns an independent number to each sow. When a sow is identified as being in estrus, its estrus index is saved under the corresponding sow number.
[0021] Beneficial effects: An image recognition database is established for each sow, and the facial features of each sow are stored in the database. For subsequent use, the facial features of the sow in the real-time image information are indexed in real time in the database, and the facial features of the sow in the captured real-time image information are extracted, and then compared in the database to find the unique corresponding sow. The estrus status of the observed pig is then saved in the database of the pig, which is convenient for the breeder to check and know the estrus status of each pig.
[0022] Furthermore, the sows are classified according to the estrus index, and the sows that have the first estrus index and the second estrus index but not the third estrus index are determined to be in false estrus.
[0023] Beneficial effects: sows in false estrus do not react obviously to boars, and their tails hang down naturally or they walk with their tails tucked between their legs, which does not meet the second and third estrus indicators. That is, when a sow shows the first estrus indicator but not the second and third estrus indicators, it may be false estrus. For sows in false estrus, timely remind the breeders to pay attention to them, prevent sow reproductive diseases, and ensure the health of the sows.
[0024] Furthermore, it also includes a temperature control module, which is used to identify and adjust the temperature of the sow's living environment to a suitable temperature.
[0025] Beneficial effects: Ensure the healthy growth of sows, enable sows to grow under suitable temperature conditions, reduce the possibility of disease, improve the physique of sows, and thus ensure the health of newborn piglets. BRIEF DESCRIPTION OF THE DRAWINGS
[0026] Figure 1 A schematic diagram of a livestock statistical monitoring system. DETAILED DESCRIPTION
[0027] The following is further described in detail through specific implementation methods:
[0028] Embodiment 1 is as attached Figure 1 As shown,
[0029] A livestock statistics monitoring system includes a photoelectric switch. The photoelectric switch sensor uses light beams such as near-infrared and infrared to detect and identify objects. The weak light beam emitted instantly by the photoelectric device can be safely, reliably and accurately emitted and received. Light is an electromagnetic wave with a transmission speed of about 300,000 km / s. It is received by it at the moment of emission, so that the sensor responds quickly. The emitted light is blocked by livestock, resulting in the inability to receive it. The photoelectric switch sensor uses the number of times the light is blocked as the livestock count mark. In order to facilitate the accuracy of the counting, the height of the photoelectric switch is adjustable, and the height is set to the height of the livestock's trunk to ensure that only one counting signal can be triggered after a livestock passes. The counting signal must be updated once each time the photoelectric switch is blocked, that is, a first counting signal is generated. It also includes:
[0030] Counting module: including a counting channel, the counting channel is used to accommodate livestock passing through, the photoelectric switch is fixed inside the counting channel, the width of the counting channel is adjustable, and the maximum width of the counting channel does not exceed 2 times the width of the livestock; the counting channel can be paved by a fence, the width of the channel can be adjusted according to the size of the livestock, and multiple counting channels can be laid to speed up the counting speed. Photoelectric switches can also be set at multiple positions of the counting channel, that is, multiple groups of photoelectric switches can be set, and the statistical quantity of each photoelectric switch within a certain period of time is compared to complete the data verification, for example, photoelectric switches are installed at two different positions in the channel, and the first value is recorded after a wave of livestock passes through the first photoelectric switch, and the second value is recorded after the same wave of livestock passes through the second photoelectric switch. The first value and the second value are compared. If they are different, they are marked and re-verified through the information of the image acquisition module to ensure the accuracy of the counting. If the values of the two light spot switches are consistent after verification, the verification is successful. If there is still an error, the data and images of the relevant time period are handed over to the breeder for manual review to complete the verification. Press sensors are arranged on both sides of the counting channel. The press sensors include a press switch, a power supply, a protection resistor, and an ammeter. The power supply, the press switch, the protection resistor, and the ammeter are electrically connected in sequence.
[0031] When there is no squeezing, the pressing switch will disconnect the circuit. When squeezing occurs, the circuit will be briefly connected. After the squeezing disappears, the pressing switch returns to its original position and continues to disconnect the circuit. The electrical signal generated during this period is transmitted to the data processing module through the Bluetooth module.
[0032] Image acquisition module: The image acquisition module is used to acquire image information / video information of livestock in the counting channel; the image information / video information at least includes information before the livestock enters the photoelectric switch and when the livestock enters the photoelectric switch;
[0033] Data processing module: The data processing module is used to obtain image information / video information, cluster the pixels in the image, and extract the contour of the livestock; the data processing module analyzes the contour of the livestock near the photoelectric switch, and before the livestock enters the photoelectric switch, if the pixel points of the two livestock in front and behind intersect, and the ratio of the number of pixels at the intersection to the number of pixels of one of the livestock exceeds a threshold, a first occlusion signal is generated; when the livestock enters the photoelectric switch, if the pixel points of the two livestock in front and behind intersect, and the ratio of the number of pixels at the intersection to the number of pixels of one of the livestock exceeds a threshold, a second occlusion signal is generated; the occlusion ratio threshold is based on 10%.
[0034] Among them, the data processing module also obtains the first counting signal of the photoelectric switch in real time. When the data processing module receives the first blocking signal, the current first counting signal is locked; if the data processing module receives the second blocking signal again, and the new value of the first counting signal does not match the actual number of livestock passing through, it is determined that the counting is missing, and the original locked first counting signal is accumulated with the number of consecutive second blocking signals to form a new first counting signal.
[0035] The livestock described are sows. The symptoms of estrus in sows are manifested in both physical and behavioral aspects. The vulva of estrus sows will become red and swollen, and dark-colored mucus will flow out. Their ears will stand up and their tails will curl upwards. In terms of behavior, they will keep walking, climb pens or straddle other sows, and automatically approach boars.
[0036] The data processing module is also used to extract the contour information of the sow's ear based on the image information / video information. Through image recognition technology, if the contour of the sow's ear presents an upward pointed corner, a first estrus indicator is generated.
[0037] A boar odorant is also provided in the counting channel where the sow moves toward the photoelectric switch, and the boar odorant is used to attract the sow. After the sow is attracted by the boar odorant, the image acquisition module is also used to acquire the state characteristics of the sow, and the state characteristics include the changes in the body shape of the sow before and after being attracted. If the sow remains still for more than a threshold time after acquiring the boar odorant, a second estrus index is generated; the image acquisition module is also used to acquire the contour information of the sow's tail, and the data processing module constructs a line model with the same trend according to the contour information of the sow's tail, and establishes a coordinate system at the last turning point of the line model toward the tail tip of the sow's tail. If the tail tip is facing upward, a third estrus index is generated.
[0038] The facial feature information of each sow is extracted in advance to form a database. The data processing module assigns an independent number to each sow. When a sow is identified as being in estrus, its estrus index is saved under the corresponding sow number, which is convenient for breeding staff to identify.
[0039] False estrus, also known as pregnancy estrus, is a sow that has slight estrus performance within 1 to 2 days before and after the next scheduled estrus period after conception. This false estrus is different from true estrus. Its estrus performance is not obvious. Generally, only the vulva is slightly swollen, and the sow is slightly uneasy, but the appetite decreases, and it can lie down quietly to rest after eating; when the sow is raised, its tail is not raised and shaken like a sow in true estrus, but droops naturally or walks with its tail tucked between legs; sows in false estrus do not react obviously to boars and generally refuse to mate. Sows are classified according to estrus indicators. Sows that show the first and second estrus indicators but not the third estrus indicator are judged as false estrus.
[0040] It also includes a temperature control module, which is used to identify and adjust the temperature of the sow's living environment to a suitable temperature. The temperature control module uses an air-conditioning compressor that can display temperature. The temperature requirements of sows at different stages of production are different. For example, the suitable temperature for sow reserve and pregnant sows is 18℃-21℃. When the temperature control module detects that the indoor ambient temperature is lower or higher than this range, it starts to cool down or heat the room, so that the indoor temperature is kept constant within the suitable temperature range.
[0041] This solution combines image recognition with photoelectric switches to achieve more accurate statistical monitoring. At the same time, by further identifying the behavior of the sow in the image, it determines whether the sow is in estrus or not, saving a lot of manpower and material resources.
[0042] The above are only embodiments of the present invention. The invention is not limited to the fields involved in this implementation case. The common sense such as the known specific structures and characteristics in the scheme is not described in detail here. The ordinary technicians in the relevant field know all the common technical knowledge in the technical field to which the invention belongs before the application date or the priority date, can obtain all the existing technologies in the field, and have the ability to apply the conventional experimental means before that date. The ordinary technicians in the relevant field can improve and implement this scheme in combination with their own abilities under the enlightenment given by this application. Some typical known structures or known methods should not become obstacles for ordinary technicians in the relevant field to implement this application. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can be made, which should also be regarded as the protection scope of the present invention, which will not affect the effect of the implementation of the present invention and the practicality of the patent. The protection scope required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to explain the content of the claims.
Claims
1. A livestock statistics monitoring system, comprising a photoelectric switch, wherein the photoelectric switch is used to count livestock passing through the photoelectric switch and generate a first counting signal, characterized in that: Also includes: Counting module: comprising a counting channel, the counting channel is used to accommodate livestock passing through, the photoelectric switch is fixed inside the counting channel, the width of the counting channel is adjustable, and the maximum width of the counting channel does not exceed 2 times the width of the livestock body; Image acquisition module: The image acquisition module is used to acquire image information / video information of livestock in the counting channel; the image information / video information at least includes information before the livestock enters the photoelectric switch and when the livestock enters the photoelectric switch; Data processing module: The data processing module is used to obtain image information / video information, cluster the pixels in the image, and extract the outline of the livestock; the data processing module analyzes the outline of the livestock near the photoelectric switch, and before the livestock enters the photoelectric switch, if the pixels of the two livestock in front and behind intersect, and the ratio of the number of pixels at the intersection to the number of pixels of one of the livestock exceeds a threshold, a first blocking signal is generated; when the livestock enters the photoelectric switch, if the pixels of the two livestock in front and behind intersect, and the ratio of the number of pixels at the intersection to the number of pixels of one of the livestock exceeds a threshold, a second blocking signal is generated; The data processing module also acquires the first counting signal of the photoelectric switch in real time. When the data processing module receives the first blocking signal, the current first counting signal is locked. If the data processing module receives the second blocking signal again, and the new value of the first counting signal does not match the actual number of livestock passing through, it is determined that the counting is missing, and the original locked first counting signal is accumulated with the number of consecutive second blocking signals to form a new first counting signal. The livestock is a sow, and the data processing module is further used to extract contour information of the sow's ear according to the image information / video information, and if the contour of the sow's ear presents a sharp angle rising upward, a first estrus indicator is generated; A boar odorant is also arranged in the counting channel where the sows move toward the photoelectric switch. The boar odorant is used to attract the sows. Press sensors are arranged on both sides of the channel. The press sensors include a press switch, a power supply, a protective resistor, and an ammeter. The power supply, the press switch, the protective resistor, and the ammeter are electrically connected in sequence. After the sow is attracted by the boar odorant, the image acquisition module is also used to obtain the state characteristics of the sow, which include the changes in the sow's body shape before and after being attracted. If the sow remains still for more than a threshold time after acquiring the boar odorant, a second estrus indicator is generated; the image acquisition module is also used to obtain the contour information of the sow's tail, and the data processing module constructs a line model with the same trend based on the contour information of the sow's tail, and establishes a coordinate system at the last turning point of the line model toward the tip of the sow's tail. If the tip of the tail is facing upward, a third estrus indicator is generated.
2. The animal husbandry statistics monitoring system according to claim 1, characterized in that: The facial feature information of each sow is extracted in advance to establish a database, and the data processing module assigns an independent number to each sow. When a sow is identified as being in estrus, its estrus index is saved under the corresponding sow number.
3. The animal husbandry statistics monitoring system according to claim 2, characterized in that: Sows are classified according to their estrus indicators, and sows that show the first and second estrus indicators but not the third estrus indicator are judged to be in false estrus.
4. The animal husbandry statistics monitoring system according to claim 3, characterized in that: It also includes a temperature control module, which is used to identify and adjust the temperature of the sow's living environment to a suitable temperature.