CN115468598A - Intelligent monitoring method and system for pigsty environment - Google Patents
Intelligent monitoring method and system for pigsty environment Download PDFInfo
- Publication number
- CN115468598A CN115468598A CN202210974517.5A CN202210974517A CN115468598A CN 115468598 A CN115468598 A CN 115468598A CN 202210974517 A CN202210974517 A CN 202210974517A CN 115468598 A CN115468598 A CN 115468598A
- Authority
- CN
- China
- Prior art keywords
- pigsty
- live
- data
- time
- pig
- 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.)
- Pending
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 72
- 238000000034 method Methods 0.000 title claims abstract description 39
- 241000282887 Suidae Species 0.000 claims abstract description 77
- 238000004140 cleaning Methods 0.000 claims abstract description 25
- 238000012545 processing Methods 0.000 claims abstract description 22
- 230000006399 behavior Effects 0.000 claims abstract description 11
- 238000001514 detection method Methods 0.000 claims abstract description 5
- 238000009395 breeding Methods 0.000 claims description 14
- 230000001488 breeding effect Effects 0.000 claims description 14
- 239000002699 waste material Substances 0.000 claims description 10
- 238000012806 monitoring device Methods 0.000 claims description 8
- 238000004891 communication Methods 0.000 claims description 6
- 230000004069 differentiation Effects 0.000 claims description 4
- 230000003068 static effect Effects 0.000 claims description 4
- 230000037396 body weight Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 claims description 3
- 230000000875 corresponding effect Effects 0.000 description 43
- 238000005516 engineering process Methods 0.000 description 3
- 230000036541 health Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000036760 body temperature Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 244000144980 herd Species 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 206010063746 Accidental death Diseases 0.000 description 1
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 206010039740 Screaming Diseases 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004040 coloring Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000013011 mating Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000016087 ovulation Effects 0.000 description 1
- 230000004962 physiological condition Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01K—ANIMAL HUSBANDRY; AVICULTURE; APICULTURE; PISCICULTURE; FISHING; REARING OR BREEDING ANIMALS, NOT OTHERWISE PROVIDED FOR; NEW BREEDS OF ANIMALS
- A01K29/00—Other apparatus for animal husbandry
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Human Computer Interaction (AREA)
- Computational Linguistics (AREA)
- Environmental Sciences (AREA)
- Software Systems (AREA)
- Animal Husbandry (AREA)
- Biodiversity & Conservation Biology (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Acoustics & Sound (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a pigsty environment intelligent monitoring method and a pigsty environment intelligent monitoring system, wherein the method comprises the following steps: acquiring and processing monitoring data of a plurality of pigsties to obtain a live image data set and a live audio data set; performing image recognition processing on the live image data set and counting to obtain the number of live pigs in each pigsty; respectively calculating the residual excrement amount of each pigsty in the time T according to the number of the live pigs and preset physiological habit prior data of the live pigs, and respectively marking as a-1, a-2, \ 8230 \ 8230;, a-n; judging whether the pigsty meets the cleaning condition or not according to the a of each pigsty, if so, recording the corresponding pigsty number, and marking the current time to obtain the details to be cleaned; performing acoustic feature analysis on the live audio data set, calibrating suspected fighting behaviors to obtain a dangerous node set, and performing image recognition processing by adopting a behavior detection method based on dangerous nodes to obtain dangerous details; and summarizing all the cleaning details and the danger details as pigsty prompt information to be sent to a specified user side.
Description
Technical Field
The invention relates to the field of intelligent breeding, in particular to a pigsty environment intelligent monitoring method and system.
Background
At present, some countries, especially several developed countries, have breeding industries with the characteristics of low labor, high yield, low consumption and the like, and the key points of the development of the breeding industries are the modernization and intellectualization of farms.
With the steady development of the domestic science and technology level, the intelligent agriculture and the breeding industry are perfected and realized. Taking pig raising as an example, there are currently:
1. marking ear tags on live pigs, automatically filing in an asset management system, acquiring a unique identity through the Internet of things technology, tracking the physical positions of the pigs, ensuring that the identity of the pigs cannot be tampered, and forming a data closed loop; financial data are correlated, and the biological asset safety of pig raising enterprises is ensured;
2. the device automatically collects body temperature data of the live pigs and uploads the body temperature data to the cloud end in real time, an AI algorithm model of the cloud end can output judgment on physiological conditions of each live pig individual after data processing, and health conditions of the live pigs can be monitored timely and accurately; meanwhile, the ovulation time of the sow is evaluated and judged, and the mating success rate and the annual number born by the sow are improved;
3. the number of live pigs in the pigsty is checked and the behavior is remotely monitored based on the monitoring camera.
Use the aforesaid as the example, it more focuses on live pig health and sells the supervision, and live pig growth and growth environment lack supporting wisdom monitoring and regulation and control scheme, breed ecological circle in wisdom and lack to some extent.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide the intelligent monitoring method for the pigsty environment, the intelligent monitoring method for the pigsty environment can regulate and control the growth and the growth environment of live pigs, coordinate the live pig health supervision technology and perfect the construction of a live pig intelligent breeding ecological circle.
The second objective of the present invention is to provide an intelligent pigsty environment monitoring system.
The technical scheme for solving the technical problems is as follows:
a pigsty environment intelligent monitoring method comprises the following steps:
s1, acquiring monitoring data of a plurality of pigsties in a target breeding area; wherein the monitoring data comprises video/image data and audio data;
s2, performing static differentiation processing on the collected monitoring data to obtain a live image data set and a live audio data set which are distributed in a time sequence manner;
s3, performing image recognition processing on the live image data set by adopting checking logic based on the pig habits, and counting to obtain the number of pigs in each pigsty;
s4, calculating the amount of excrement accumulated in the pigsty within the time T according to the number of the live pigs in the pigsty and preset physiological habit prior data of the live pigs, and recording the amount as a;
s5, respectively calculating the amount of excrement accumulated in each pigsty within the time T, and respectively marking as a-1, a-2, \8230 \ a-n, wherein n is a natural number and represents a pigsty number;
s6, judging whether the corresponding pigsty meets the cleaning condition or not according to the amount of excrement accumulated in each pigsty within the time T, if so, recording the number of the corresponding pigsty, and marking the current time to obtain the detail to be cleaned;
s7, summarizing all the cleaning details as pigsty prompt information and sending the pigsty prompt information to a specified user side;
s8, performing acoustic feature analysis on the live audio data set based on the traversal logic, and calibrating suspected fighting behaviors to obtain a dangerous node set;
s9, according to the dangerous node set, performing image recognition processing by adopting a behavior detection method based on dangerous nodes to obtain dangerous details;
and S10, summarizing each danger detail as pigsty prompt information and sending the information to a specified user side.
Preferably, in step S3, the pig habit based inventory logic includes the following steps:
s301, extracting the image at the time t1 for feature recognition, and evaluating whether the pig is a live pig or not based on the features;
s302, judging whether the live pigs are close to each other or not based on the live pig identification data, and if yes, skipping corresponding targets; otherwise, marking and tracking are carried out;
s303, taking t as a minimum time length unit, and letting t1= t1+ t;
s304, repeating the steps S301-S303, and if the previous marked target is marked again, obtaining a corresponding number of times attribute value of +1;
and S305, when all the live pigs in the same piggery are marked, stopping circulation, and counting the marked quantity to finish inventory.
Preferably, in step S303, a value of the minimum time length unit t is:
s3031, taking the previous N times of checking records of the live pigs;
s3032, counting the target quantity of the past newly added marks and the number attribute +1 in each live pig inventory record, and recording the target quantity as X;
s3033, calculating the ratio Y of the X in the corresponding pigsty according to the corresponding checking result of the live pigs;
s3034, if Y is larger than a preset threshold value, a marking cycle corresponding to X is marked as an active node, and the number of the active nodes is marked as d; and calculating an active time length S:
S＝(d-1)×t p ；
wherein, t p The minimum time length unit used for the last live pig inventory record is recorded;
s3035, accumulating the activity duration S corresponding to the N times of live pig inventory records, and calculating the average value
S3036, according to the mean value
Finding a predetermined mean value
And obtaining the t of the next live pig inventory by taking the value relation table.
Preferably, in step S3, feeding characteristic recognition is performed on the live image data set, and a feeding start time t2 and a feeding end time t3 corresponding to the pigsty are recorded; wherein, the value of t1 in the inventory logic is outside the interval (t 2, t 3).
Preferably, in step S4, calculating the amount a of excrement accumulated in the pigsty within the time T, and calling corresponding prior waste discharge data, wherein the calling method of the prior waste discharge data is as follows;
s401, acquiring corresponding live pig affiliated parameters of each pigsty, wherein the live pig affiliated parameters comprise the latest time of entering a pigsty and the weight of the live pig measured before the latest time of entering the pigsty;
s402, importing the auxiliary parameters of the live pigs into a preset live pig growth prediction model to obtain real-time estimated weight;
and S403, searching the prior data of the physiological habits of the live pigs according to the estimated body weight to obtain the corresponding prior waste discharge data.
Preferably, in step S8, in step (6), the cleaning conditions are: whether the accumulated excrement amount a in the time T is larger than a set threshold value and whether the accumulated excrement amount a meets the requirement that the time T is within the cleaning time in the work plan schedule of the staff or not; when the two conditions are met simultaneously, the cleaning condition is met.
Preferably, in step S8, the traversal logic includes:
s801, extracting audio in a time period from t1 to t2 for feature identification, and evaluating whether the pig is a live pig or not based on features;
s802, judging whether live pig fighting conditions are met or not based on the acoustic characteristics of the live pigs, and skipping the current audio if the live pig fighting conditions are not met; if yes, recording a corresponding time period as a risk node;
s803, with Δ t as a time length increment unit, and let t1= t2, t2= t1+ t;
and S804, repeating the steps S801 to S803 until the audio data is completely traversed to obtain a dangerous node set.
Preferably, before executing the traversal logic, audio data fed back by a plurality of audio acquisition devices pre-installed in the pigsty are acquired; acquiring and identifying the equipment ID of the audio acquisition device corresponding to the audio data; determining a corresponding audio acquisition device according to the audio files meeting the live pig fighting conditions; and searching attributive monitoring data according to the pre-recorded position data of the audio acquisition device.
Preferably, in step S9, the behavior detection method based on the dangerous node includes the following steps:
s901, calling a corresponding image in the live image data set according to the risk node, and recording the image as a risk image;
s902, carrying out feature identification on the risk image, and evaluating whether the risk image is a live pig or not based on features;
s903, judging whether the live pig is close to the target or not based on the live pig identification data, and if so, skipping the corresponding target; if not, marking;
s904, calculating the number of the close live pigs according to the live pig inventory result and the marked amount;
s905, judging whether the number of the close live pigs is larger than a preset risk threshold value or not, and if so, adding the risk pictures into the risk details.
An intelligent pigsty environment monitoring system comprises a monitoring device, a communication device and a central processing device, wherein the monitoring device is used for acquiring monitoring data of a plurality of pigsties in a target breeding area and comprises an image acquisition device and an audio acquisition device; the communication device is used for uploading the collected monitoring data to the central processing device, the central processing device comprises a memory and a processor, the memory is used for storing data information including the monitoring data, and the processor loads the data information in the memory and executes the data information according to the intelligent pig house environment monitoring method.
Compared with the prior art, the invention has the following beneficial effects:
1. according to the intelligent monitoring method for the pigsty environment, the image data are processed through detecting the image data, so that the live pigs in each pigsty are checked, and the situations that the live pigs escape, climb over and enter other pigsties and the like are found in time.
2. According to the intelligent monitoring method for the pigsty environment, the accumulated excrement amount of the pigsty in the time T can be calculated according to the number of the live pigs in the pigsty and the preset physiological habit prior data of the live pigs, and if the cleaning condition is met, relevant workers are informed timely, so that the workers can clean the excrement timely to guarantee the pigsty environment.
3. According to the intelligent monitoring method for the pigsty environment, whether live pigs are placed in the pigsty can be determined through audio matching of the pigsty and the monitoring data and cross comparison, so that the situations of accidental death of the live pigs and the like are reduced, and the economic loss is reduced.
4. The intelligent pigsty environment monitoring system can automatically check the live pigs in each pigsty, simultaneously monitor the excrement in each pigsty, and timely inform workers of cleaning when cleaning conditions are met, so that the pigsty environment is ensured; in addition, the pig behavior monitoring device can monitor the pig behaviors, prevent the pigs from being placed on a shelf, and can quickly position the placed pigs when the pigs are placed on the shelf, so that workers can conveniently evacuate the pigs in time, and casualties of the pigs are reduced.
Drawings
Fig. 1 is a schematic view illustrating the intelligent monitoring system for pigsty environment according to the present invention.
Fig. 2 is a schematic main flow chart of a first embodiment of the intelligent pigsty environment monitoring method of the invention.
Fig. 3 is a schematic main flow chart of a pigsty environment intelligent monitoring method according to a second embodiment of the invention.
Detailed Description
The present invention will be described in further detail with reference to examples and drawings, but the present invention is not limited thereto.
Example 1
Referring to fig. 1-2, the intelligent monitoring method for pigsty environment of the invention comprises the following steps:
s1, acquiring monitoring data of a target breeding area; wherein a plurality of pigsties are arranged in the target breeding area;
the monitoring data are shot and collected through a monitoring device (such as a monitoring camera) arranged in the target breeding area and uploaded through a network. In order to aim at subsequent target characteristics and requirements, the monitoring camera is installed on a ceiling support above the pigsty, and the shooting visual angle of the monitoring camera is mainly overlooked.
In this example, a plurality of types of live pigs were mixed and housed in the pig house.
S2, performing static differentiation processing on the monitoring data to obtain a time sequence distributed live image data set;
wherein the purpose of the static differentiation treatment is to obtain an image; if the monitoring data takes video data as an example, extracting an image of a certain frame at a specified frequency; if the monitoring data is not a video but a frequency snapshot type, analyzing the definition of the image, and then taking the image of which the definition meets a preset standard.
S3, performing image recognition processing on the live image data set by using checking logic based on the pig habits, and counting to obtain the number of pigs in each pigsty;
wherein, the inventory logic specifically includes:
s301, extracting the image at the time t1 for feature recognition, and evaluating whether the pig is a live pig or not based on the features;
in the embodiment, the feature extraction in the image recognition is identified as the prior art, and therefore, the description is not repeated; for live pigs, the applicable characteristics are pig heads, and the identification accuracy can be guaranteed by taking the pig heads as the characteristics for extraction;
s302, judging whether the live pigs are close to each other or not based on the live pig identification data, and if so, skipping corresponding targets; if not, marking and tracking;
wherein, whether the live pig is close to the live pig can be judged by the following two ways:
(1) Calculating pixel distances between image features;
(2) Analyzing the image connected domain in the image recognition process, and determining that the connected domain of a certain image is obviously higher than other independent live pigs (separated), so as to be close;
the above-mentioned mark can be made in a coloring manner, i.e. randomly assigned with a color.
S303, taking t as a minimum time length unit, and letting t1= t1+ t;
s304, repeating the steps S301-S303, and if the previous marking target is marked again, obtaining a corresponding number of times attribute value +1;
s305, when all the live pigs in the same pigsty are marked, stopping circulation, and counting the marked quantity to finish checking;
by adopting the checking logic, the number of the live pigs in each pigsty can be input without manual checking; simultaneously, can check once live pig quantity by the automatic check of a period of time every based on the control data to in time discover that the live pig is stolen, escapes, or from a pig house the condition of crossing to another pig house, and then help the supervision of staff to the district of raising pigs.
In the above inventory logic, that is, in step S303, a value of the minimum time length unit t is:
s3031, taking the previous N times of checking records of the live pigs;
s3032, counting the target quantity of the newly added marks and the times attribute +1 in each live pig inventory record, and recording the target quantity as X;
s3033, calculating the proportion Y of the X in the corresponding pigsty according to the corresponding pig checking result;
s3034, if Y is larger than a preset threshold value, one marking cycle corresponding to X is marked as an active node, and the number of the active nodes is marked as d; and calculating the active time length S:
S＝(d-1)×t p ；
wherein, t p The minimum time length unit used for the last live pig inventory record is recorded;
s3035, accumulating the activity duration S corresponding to the N times of live pig inventory records, and calculating the average value
S3036, according to the mean value
Find a preset mean
Obtaining t of the next live pig inventory by taking a value relation table, wherein the preset mean value
The value-taking relation table is obtained by an experience formula of workers based on the habits of the live pigs; for example, from a large number of data, a large number of t values and corresponding mean values are determined
It is then analyzed (e.g., linear analysis) to produce a mean value
And (5) a value relation table.
Therefore, t used in the checking process is not a fixed value, and is adjusted in a targeted mode based on the change of the live pig active time of the passing piggery, so that the checking work can be accurately carried out in the live pig active time, the interference caused by the live pigs gathering close to rest is reduced, and the cycle number and the analysis amount are further reduced.
It should be noted that inventorying is not necessary at this stage, since there is a high probability that live pigs will gather in the trough during the feeding stage; meanwhile, if the feeding stage is not eliminated, the value of t will cause great interference, thus resulting in too high distortion of t.
In step S3, feeding characteristic recognition is carried out on the live image data set, and feeding starting time t2 and feeding ending time t3 of the corresponding pigsty are recorded; wherein, the value of t1 in the inventory logic is outside the interval (t 2, t 3).
S4, calculating the amount of excrement accumulated in the pigsty within the time T according to the number of the live pigs in the pigsty and preset physiological habit prior data of the live pigs, and recording the amount as a;
the said physiological habit prior data of live pig refers to: the worker verifies the obtained waste discharge data of each growth cycle (weight division) of the live pigs by measuring the weights of a plurality of live pigs and daily excrement amount.
In step S4, calculating the amount a of excrement accumulated in the pigsty within the time T, and calling corresponding prior waste discharge data, wherein the calling method of the prior waste discharge data is as follows;
s401, acquiring affiliated parameters of the live pigs corresponding to the piggeries; the auxiliary parameters comprise (manually) recorded weight and entering time measured before entering the column for the latest time;
s402, importing the auxiliary parameters of the live pigs into a preset live pig growth prediction model to obtain real-time estimated weight;
and S403, searching the priori data of the physiological habits of the live pigs according to the estimated body weights to obtain the corresponding priori waste discharge data.
The live pig growth prediction model in the step S402 can be obtained by acquiring a large amount of live pig growth time and corresponding weight by workers and then introducing the live pig growth time and the corresponding weight into a neural network model for training; the establishment and training of the model are prior art and thus are not described in detail.
S5, respectively calculating the amount of excrement accumulated in each pigsty within the time T, and respectively marking as a-1, a-2, 8230, a-n; wherein n is a natural number and represents the number and the identification code of the pigsty.
S6, judging whether the corresponding pigsty meets the cleaning condition or not according to the amount of excrement accumulated in each pigsty within the time T, if so, recording the number of the corresponding pigsty, and marking the current time to obtain the detail to be cleaned;
the above-mentioned cleaning condition may be a threshold value corresponding to the amount of excrement a; more effectively, the work plan schedule of the relevant workers in the culture area is imported, whether the work plan schedule belongs to the cleaning time or not is judged, and if the work plan schedule belongs to the cleaning time and the previous threshold condition is met, the work plan schedule accords with the cleaning condition; namely, the accumulated excrement amount a (n) in the time T is larger than a set threshold value, and the cleaning condition is met when the time T is met within the cleaning time in the work plan schedule of the worker.
And S7, summarizing all the cleaning details as pigsty prompt information and sending the pigsty prompt information to a specified user side, wherein all the cleaning details are summarized to form a detail list which can be sent to the client side through a short message, a matched APP and a notification prompt of a UI interactive interface.
By the mode, the intelligent monitoring method for the pigsty environment can count the number of the live pigs in each pigsty based on the monitoring data, and timely find the situations that the live pigs escape, climb over and enter other pigsties and the like; meanwhile, the excrement cleaning time of each pigsty can be predicted, and related workers can be informed in time so as to guarantee the pigsty environment. After the staff finishes the clearance of pig house once, need to pass through the information such as personal terminal with pig house clearance time upload, accomplish the zero clearing that resets of pig house long-pending excrement volume to the prediction of clearance time next time.
In another embodiment of the intelligent pigsty environment monitoring method, the situation that the live pigs die accidentally and the like is reduced and economic loss is reduced by acquiring the audio data of the pigsty and matching the audio data with the monitoring data to determine whether the live pigs are placed in the pigsty through cross comparison; the method comprises the following specific steps:
(1) Acquiring audio data fed back by a plurality of audio acquisition devices (namely sound pickups) pre-installed in the pigsty;
(2) Performing acoustic feature identification on the audio data (for example, decibels and energy levels are high and low, a live pig with a sound will make a sharp call), judging whether live pig putting conditions are met (for example, judging whether corresponding features exceed a threshold value and whether the type of the sound meets the requirements), and if so, recording corresponding time as a risk node;
(3) Calling a corresponding image in the live image data set according to the risk node, and recording the image as a risk image;
(4) Performing feature identification on the risk image, and evaluating whether the risk image is a live pig or not based on the features;
(5) Judging whether the live pigs are close to each other or not based on the live pig identification data, and if so, skipping corresponding targets; if not, marking;
(6) Calculating the number of the close live pigs according to the live pig inventory result and the marked amount;
(7) And judging whether the number of the close live pigs is larger than a preset risk threshold value or not, and if so, sending a risk picture to a specified user side.
That is, when the screaming with high decibel and ear pricking occurs in the pigsty, whether the risk of putting up the pig is present or not is judged according to the audio data, if so, the image recognition and analysis of the corresponding time is carried out, and if the number of the adjacent pigs is large (the pig herd is put up, so that other pigs in the same pigsty are close to each other and are active), the corresponding image is sent to the staff, so that the staff can find out the abnormality of the pigsty in time. The rest of the pig herd is not only too little to feed, but also the environment of the pigsty is deteriorated, and the number of the live pigs in the same pigsty is too large; by the method, the problem can be easily found by workers, so that the problem can be solved in time, the death probability of the live pigs can be effectively reduced, and the economic loss can be reduced.
In order to reduce the data processing amount in the live pig shelving analysis process and improve the accuracy, the intelligent monitoring method for the pigsty environment is improved as follows:
(101) Acquiring and recording effective shooting range data of each monitoring data (such as pigsties with numbers of 1-n, preferably one to two pigsties correspond to one camera);
(102) Acquiring and identifying the equipment ID of the audio acquisition device corresponding to the audio data;
(103) Determining a corresponding audio acquisition device according to the audio file meeting the live pig fighting condition;
(104) Searching attributive monitoring data according to the pre-recorded position data of the audio acquisition device (namely, if the audio acquired by the sound pickup in which area is abnormal, subsequently calling the monitoring data in the area);
(105) And retrieving images from the live graph dataset according to the corresponding monitoring data and risk nodes.
Through the improvement, the intelligent monitoring method for the pigsty environment accurately searches and calls data layer by layer for live image data sets distributed in the time sequence of the breeding area through the one-to-one corresponding mapping relation, so that the analysis rate is effectively improved.
Example 2
The intelligent pigsty environment monitoring system comprises a monitoring device, a communication device and a central processing device, wherein the monitoring device is used for acquiring monitoring data of a plurality of pigsties in a target breeding area and comprises an image acquisition device (such as a camera) and an audio acquisition device (such as a sound pickup); the communication device is used for uploading the collected monitoring data to a central processing device, the central processing device comprises a memory and a processor, the memory is used for storing data information including the monitoring data, and the processor loads the data information in the memory and executes the data information according to the computer program of the intelligent piggery environment monitoring method.
The present invention is not limited to the above embodiments, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be construed as equivalents and are included in the scope of the present invention.
Claims (10)
1. An intelligent pigsty environment monitoring method is characterized by comprising the following steps:
s1, acquiring monitoring data of a plurality of pigsties in a target breeding area; wherein the monitoring data comprises video/image data and audio data;
s2, performing static differentiation processing on the collected monitoring data to obtain a live image data set and a live audio data set which are distributed in a time sequence;
s3, performing image recognition processing on the live image data set by adopting checking logic based on the pig habits, and counting to obtain the number of pigs in each pigsty;
s4, calculating the amount of excrement accumulated in the pigsty within the time T according to the number of the live pigs in the pigsty and preset physiological habit prior data of the live pigs, and recording the amount as a;
s5, respectively calculating the amount of excrement accumulated in each pigsty within the time T, and respectively marking as a-1, a-2, \8230 \ a-n, wherein n is a natural number and represents a pigsty number;
s6, judging whether the corresponding pigsty meets the cleaning condition or not according to the amount of excrement accumulated in each pigsty within the time T, if so, recording the number of the corresponding pigsty, and marking the current time to obtain the detail to be cleaned;
s7, summarizing all the cleaning details as pigsty prompt information and sending the pigsty prompt information to a specified user side;
s8, performing acoustic feature analysis on the live audio data set based on traversal logic, and calibrating suspected fighting behaviors to obtain a dangerous node set;
s9, according to the dangerous node set, performing image recognition processing by adopting a behavior detection method based on dangerous nodes to obtain dangerous details;
and S10, summarizing each danger detail as piggery prompt information and sending the piggery prompt information to a specified user side.
2. The intelligent pigsty environment monitoring method according to claim 1, wherein in step S3, the checking logic based on the habits of the live pigs comprises the following steps:
s301, extracting the image at the time t1 for feature recognition, and evaluating whether the pig is a live pig or not based on the features;
s302, judging whether the live pigs are close to each other or not based on the live pig identification data, and if so, skipping the corresponding target; otherwise, marking and tracking are carried out;
s303, taking t as a minimum time length unit, and letting t1= t1+ t;
s304, repeating the steps S301-S303, and if the previous marking target is marked again, obtaining a corresponding number of times attribute value +1;
and S305, when all the pigs in the same piggery are marked, stopping circulation, and counting the marked quantity to finish inventory.
3. The intelligent pigsty environment monitoring method according to claim 2, wherein in step S303, the minimum time length unit t is obtained by:
s3031, taking the previous checking records of the live pigs for N times;
s3032, counting the target quantity of the newly added marks and the times attribute +1 in each live pig inventory record, and recording the target quantity as X;
s3033, calculating the proportion Y of the X in the corresponding pigsty according to the corresponding pig checking result;
s3034, if Y is larger than a preset threshold value, one marking cycle corresponding to X is marked as an active node, and the number of the active nodes is marked as d; and calculating an active time length S:
S＝(d-1)×t p ；
wherein, t p The minimum time length unit used for the last live pig inventory record;
s3035, accumulating the activity duration S corresponding to the N times of live pig inventory records, and calculating the average value
S3036, according to the mean value
Find a preset mean
And obtaining the t of the next live pig inventory by taking the value relation table.
4. The intelligent pigsty environment monitoring method according to claim 3, wherein in step S3, feeding characteristic identification is performed on the live image data set, and feeding start time t2 and feeding end time t3 corresponding to the pigsty are recorded; wherein, the value of t1 in the inventory logic is outside the interval (t 2, t 3).
5. The intelligent pigsty environment monitoring method according to claim 1, wherein in step S4, calculating the amount a of excrement accumulated in the pigsty within the time T requires calling corresponding prior waste discharge data, wherein the calling method of the prior waste discharge data is;
s401, acquiring corresponding live pig affiliated parameters of each pigsty, wherein the live pig affiliated parameters comprise the latest time of entering a pigsty and the weight of the live pig measured before the latest time of entering the pigsty;
s402, importing the auxiliary parameters of the live pigs into a preset live pig growth prediction model to obtain real-time estimated weight;
and S403, searching the prior data of the physiological habits of the live pigs according to the estimated body weight to obtain the corresponding prior waste discharge data.
6. The intelligent pigsty environment monitoring method according to claim 1, wherein in step S8, in step (6), the cleaning conditions are: whether the accumulated excrement amount a in the time T is larger than a set threshold value and whether the accumulated excrement amount a meets the requirement that the time T is within the cleaning time in the work plan schedule of the staff or not; when the two conditions are simultaneously met, the cleaning condition is met.
7. The intelligent pig house environment monitoring method according to claim 4, wherein in step S8, the traversal logic includes:
s801, extracting audio in a time period from t1 to t2 for feature recognition, and evaluating whether the pig is a live pig or not based on features;
s802, judging whether live pig fighting conditions are met or not based on the acoustic characteristics of the live pigs, and skipping the current audio if the live pig fighting conditions are not met; if so, recording a corresponding time period as a risk node;
s803, with Δ t as a time length increment unit, and let t1= t2, t2= t1+ t;
and S804, repeating the steps S801 to S803 until the audio data is completely traversed to obtain a dangerous node set.
8. The intelligent pigsty environment monitoring method according to claim 7, wherein before executing the traversal logic, audio data fed back by a plurality of audio acquisition devices pre-installed in the pigsty are obtained; acquiring and identifying the equipment ID of the audio acquisition device corresponding to the audio data; determining a corresponding audio acquisition device according to the audio files meeting the live pig fighting conditions; and searching attributive monitoring data according to the pre-recorded position data of the audio acquisition device.
9. The intelligent pigsty environment monitoring method according to claim 1, wherein in step S9, the behavior detection method based on the danger node comprises the following steps:
s901, calling a corresponding image in the live image data set according to the risk node, and recording the image as a risk image;
s902, carrying out feature identification on the risk image, and evaluating whether the risk image is a live pig or not based on features;
s903, judging whether the live pigs are close to each other or not based on the live pig identification data, and if yes, skipping corresponding targets; if not, marking;
s904, calculating the number of the close live pigs according to the live pig inventory result and the marked quantity;
and S905, judging whether the number of the close live pigs is larger than a preset risk threshold value or not, and if so, adding the risk pictures into the risk detail.
10. A pigsty environment intelligent monitoring system adopting the pigsty environment intelligent monitoring method of any one of claims 1 to 9, which is characterized by comprising a monitoring device, a communication device and a central processing device, wherein the monitoring device is used for acquiring monitoring data of a plurality of pigsties in a target breeding area and comprises an image acquisition device and an audio acquisition device; the communication device is used for uploading the collected monitoring data to a central processing device, the central processing device comprises a memory and a processor, wherein the memory is used for storing data information including the monitoring data, and the processor loads the data information in the memory and executes the data information according to the intelligent pigsty environment monitoring method of any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210974517.5A CN115468598A (en) | 2022-08-15 | 2022-08-15 | Intelligent monitoring method and system for pigsty environment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210974517.5A CN115468598A (en) | 2022-08-15 | 2022-08-15 | Intelligent monitoring method and system for pigsty environment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115468598A true CN115468598A (en) | 2022-12-13 |
Family
ID=84368037
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210974517.5A Pending CN115468598A (en) | 2022-08-15 | 2022-08-15 | Intelligent monitoring method and system for pigsty environment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115468598A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115826654A (en) * | 2023-01-05 | 2023-03-21 | 盐城工学院 | AI-based multifunctional Internet of things monitoring system and method for pigsty |
| CN118038366A (en) * | 2024-02-20 | 2024-05-14 | 青岛法牧机械有限公司 | Intelligent monitoring system and method for pig farm cultivation |
-
2022
- 2022-08-15 CN CN202210974517.5A patent/CN115468598A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115826654A (en) * | 2023-01-05 | 2023-03-21 | 盐城工学院 | AI-based multifunctional Internet of things monitoring system and method for pigsty |
| CN118038366A (en) * | 2024-02-20 | 2024-05-14 | 青岛法牧机械有限公司 | Intelligent monitoring system and method for pig farm cultivation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109840549B (en) | Method and device for identifying plant diseases and insect pests | |
| CN115468598A (en) | Intelligent monitoring method and system for pigsty environment | |
| O’Connor et al. | Camera trap arrays improve detection probability of wildlife: Investigating study design considerations using an empirical dataset | |
| Berckmans | Precision livestock farming technologies for welfare management in intensive livestock systems | |
| CN111460990B (en) | Big data-based alpine pastoral area grassland insect pest monitoring and early warning system and method | |
| US20130006065A1 (en) | System and methods for health monitoring of anonymous animals in livestock groups | |
| Berckmans et al. | Animal sound… talks! Real-time sound analysis for health monitoring in livestock | |
| CN104077550B (en) | The method and system that a kind of health index for realizing animal behavior monitoring is evaluated | |
| CN109637549A (en) | A kind of pair of pig carries out the method, apparatus and detection system of sound detection | |
| Adriaens et al. | Milk losses and dynamics during perturbations in dairy cows differ with parity and lactation stage | |
| CN110991222B (en) | Object state monitoring and sow oestrus monitoring method, device and system | |
| CN112150498B (en) | Method and device for determining body state information, storage medium and electronic device | |
| CN115250952A (en) | Live pig health monitoring method, device, equipment and storage medium | |
| CN112906734A (en) | Intelligent livestock breeding method and device, computer equipment and storage medium | |
| KR102518418B1 (en) | Apparatus and method for analyzing ruminant and breeding environment based on image analysis | |
| CN111260496A (en) | Livestock and poultry monitoring method and system | |
| CN113516139A (en) | Data processing method, device, equipment and storage medium | |
| EP3541173B1 (en) | A method for customized monitoring of sounds caused by respiratory distress | |
| Ojukwu et al. | Development of a computer vision system to detect inactivity in group-housed pigs | |
| CN114532248B (en) | Heat stress behavior monitoring method and monitoring device for dairy cow | |
| CN113792702A (en) | Video streaming livestock monitoring method and system based on computer vision | |
| McGrath et al. | Recent spatial changes in bovine tuberculosis in the Republic of Ireland | |
| CN112560750A (en) | Video-based ground cleanliness recognition algorithm | |
| CN118155144B (en) | Vegetable planting pesticide input supervision system and method based on AI vision | |
| US20240074412A1 (en) | System and method for analyzing meting behavior of an animal species |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |