CN-117218577-B - Meat pigeon behavior counting method and device, electronic equipment and storage medium

CN117218577BCN 117218577 BCN117218577 BCN 117218577BCN-117218577-B

Abstract

The invention relates to the technical field of data processing, in particular to a pigeon behavior counting method, electronic equipment and a storage medium, wherein the pigeon behavior counting method obtains the ingestion behavior of a target pigeon in each frame of a video stream by identifying the video stream obtained by shooting the pigeon in real time frame by frame, acquires the frame numbers of continuous frames with the same ingestion behavior, the feeding times corresponding to the continuous frame number are obtained by combining the counting model, so that the automatic identification and counting of the pigeon behaviors are realized, accurate feeding can be performed according to the feeding times of the pigeons, the unreasonable feeding phenomenon is reduced, the breeding efficiency and quality are improved, and the labor cost and the management cost are reduced.

Inventors

FENG DACHUN
LIU SHUANGYIN
XU LONGQIN
XIE JIEFENG
WANG YONGKANG
LIU TONGLAI
CHEN WEIBO

Assignees

仲恺农业工程学院
梅州市金绿现代农业发展有限公司

Dates

Publication Date: 20260505
Application Date: 20230904

Claims (8)

1. The pigeon behavior counting method is characterized by comprising the following steps of: acquiring a video stream obtained by shooting meat pigeons in real time; the method comprises the steps of carrying out frame-by-frame identification on a video stream by adopting a preset target detection model to obtain ingestion behaviors of target pigeons in frames of the video stream, wherein the target detection model comprises a backbone network, a feature fusion network and a detection head, and comprises the steps of carrying out feature extraction on each frame of image of the video stream through the backbone network to obtain multi-dimensional features corresponding to each frame of image of the video stream; Acquiring the number of continuous frames according to the ingestion behaviors, wherein the number of the continuous frames is the number of the continuous frames with the same ingestion behaviors; Obtaining ingestion times corresponding to the continuous frame number according to the continuous frame number and a constructed counting model, wherein the counting model describes the corresponding relation between the continuous frame number and the ingestion times, and the construction of the counting model comprises the steps of carrying out regression modeling on the corresponding relation between the continuous frame number and the ingestion times by adopting a four-parameter classification variable regression function to obtain the counting model; The expression of the four parameter Probit function is as follows: Wherein, the Representing the number of consecutive frames, being an argument of a four parameter Probit function; representing the ingestion times, is a dependent variable of a four-parameter Probit function, and the parameters a, b, c and d are variable parameters obtained by model fitting, wherein And A represents the growth rate of the function, b represents the inflection point of the function; A cumulative distribution function representing a normal distribution, the function expressed as: When (when) The feeding frequency becomes slow with the increase of the frame number, and finally approaches to the upper asymptote 。
2. The pigeon behavior counting method according to claim 1, wherein the step of identifying the video stream frame by using a preset target detection model to obtain the feeding behavior of the target pigeon in each frame of the video stream comprises the steps of: and carrying out frame-by-frame identification on the video stream within a preset time period by adopting the target detection model to obtain the ingestion behavior.
3. The pigeon behavior counting method according to claim 1, wherein the counting method further comprises: identifying the gender of the target meat pigeon; Correlating said gender with said feeding behavior and said number of feeding.
4. A pigeon behavioral counting method according to claim 3, wherein said identifying the sex of the target pigeon comprises: And identifying the sex of the target pigeon according to the video stream and the sex identification preset on the target pigeon.
5. The pigeon behavior counting method according to claim 1, wherein the counting method further comprises: and storing the video stream, the ingestion behaviors in each frame of the video stream and the ingestion times corresponding to the ingestion behaviors.
6. A pigeon behavior counting device, comprising: the first acquisition module is used for acquiring a video stream obtained by shooting the meat pigeons in real time; The identification module is used for carrying out frame-by-frame identification on the video stream by adopting a preset target detection model to obtain the ingestion behavior of the target pigeon in each frame of the video stream, wherein the target detection model comprises a backbone network, a feature fusion network and a detection head, and comprises the steps of carrying out feature extraction on each frame of image of the video stream through the backbone network to obtain the multi-dimensional feature corresponding to each frame of image of the video stream; The second acquisition module is used for acquiring the number of continuous frames according to the ingestion behaviors, wherein the number of the continuous frames is the number of the continuous frames with the same ingestion behaviors; the counting module is used for obtaining the ingestion times corresponding to the continuous frame number according to the continuous frame number and a constructed counting model, wherein the counting model describes the corresponding relation between the continuous frame number and the ingestion times, and the construction of the counting model comprises the steps of carrying out regression modeling on the corresponding relation between the continuous frame number and the ingestion times by adopting a four-parameter classification variable regression function to obtain the counting model; The expression of the four parameter Probit function is as follows: Wherein, the Representing the number of consecutive frames, being an argument of a four parameter Probit function; representing the ingestion times, is a dependent variable of a four-parameter Probit function, and the parameters a, b, c and d are variable parameters obtained by model fitting, wherein And A represents the growth rate of the function, b represents the inflection point of the function; A cumulative distribution function representing a normal distribution, the function expressed as: When (when) The feeding frequency becomes slow with the increase of the frame number, and finally approaches to the upper asymptote 。
7. An electronic device, comprising: At least one memory and at least one processor; the memory is used for storing one or more programs; the one or more programs, when executed by the at least one processor, cause the at least one processor to implement the steps of the pigeon behavior counting method of any of claims 1-5.
8. A storage medium having stored therein a computer program which when executed by a processor performs the steps of the pigeon behavior counting method according to any one of claims 1-5.

Description

Meat pigeon behavior counting method and device, electronic equipment and storage medium Technical Field The invention relates to the technical field of data processing, in particular to a pigeon behavior counting method, a pigeon behavior counting device, electronic equipment and a storage medium. Background Along with the continuous acceleration of the urban process and the improvement of the living standard of people, the quality requirements of people on foods are becoming higher and higher, and healthy, environment-friendly and safe foods are becoming necessities in daily life of people. Meat pigeons are also becoming increasingly popular in the marketplace as a meat product with rich nutrition, delicious taste and cholesterol deficiency. However, the current lagged feeding modes and feeding technologies of meat pigeons greatly restrict the growth of meat pigeons. Firstly, most breeders can feed the pigeon feed according to own experience, and the phenomena of insufficient feed and excessive feed are easy to occur, so that the conditions of feed pollution, mildew and the like are caused, and the normal generation of the pigeon is influenced. Although a part of workers can feed meat pigeons according to feeding behaviors of the meat pigeons, manual observation of the meat pigeon behaviors not only consumes labor, time and cost, but also cannot accurately grasp the feeding amount and timing of the meat pigeons, and the problems also occur. Disclosure of Invention In order to solve the problems in the prior art, the invention provides the pigeon behavior counting method, the device, the electronic equipment and the storage medium, and realizes automatic identification and counting of pigeon behaviors, thereby realizing accurate feeding, improving the breeding efficiency and quality and reducing the cost. In order to achieve the technical purpose, the technical scheme adopted by the embodiment of the application comprises the following steps: in a first aspect, an embodiment of the present application provides a pigeon behavior counting method, including the following steps: acquiring a video stream obtained by shooting meat pigeons in real time; Carrying out frame-by-frame identification on the video stream by adopting a preset target detection model to obtain the ingestion behavior of the target pigeons in each frame of the video stream; acquiring the number of continuous frames according to the ingestion behavior, wherein the number of the continuous frames is the number of the continuous frames with the same ingestion behavior; and obtaining the ingestion times corresponding to the continuous frame number according to the continuous frame number and a constructed counting model, wherein the counting model describes the corresponding relation between the continuous frame number and the ingestion times. In addition, the pigeon behavior counting method according to the embodiment of the application may further have the following additional technical features: Further, in the pigeon behavior counting method of the embodiment of the application, the target detection model comprises a backbone network, a feature fusion network and a detection head; Adopting a preset target detection model to identify the video stream frame by frame to obtain the ingestion behavior of the target pigeons in each frame of the video stream, comprising the following steps: extracting features of each frame of image of the video stream through a backbone network to obtain multi-dimensional features corresponding to each frame of image of the video stream; Feature fusion is carried out on multidimensional features corresponding to each frame of image of the video stream through a feature fusion network, so that fusion features corresponding to each frame of image of the video stream are obtained; and obtaining feeding behaviors through the detection head according to the fusion characteristics. Further, in an embodiment of the present application, a preset target detection model is adopted to identify a video stream frame by frame, so as to obtain ingestion behaviors of a target pigeon in each frame of the video stream, including: and carrying out frame-by-frame identification on the video stream within a preset time period by adopting a target detection model to obtain the ingestion behavior. Further, in one embodiment of the present application, the construction of the counting model includes: Carrying out regression modeling on the corresponding relation between the continuous frame number and the ingestion times by adopting a four-parameter classification variable regression function to obtain a counting model; and performing model fitting on the counting model to obtain each parameter of the counting model. Further, in one embodiment of the present application, the counting method further includes: identifying the sex of the target pigeon; Gender is related to feeding behavior and number of feeding. Further, in one embodiment