CN-122023236-A - Egg quality detection method and related equipment

CN122023236ACN 122023236 ACN122023236 ACN 122023236ACN-122023236-A

Abstract

The application discloses an egg quality detection method and related equipment, belonging to the technical field of image detection, wherein the method comprises the steps of obtaining an egg image of a target area; the method comprises the steps of preprocessing an egg image to obtain preprocessed egg image data characteristics, inputting the egg image data characteristics into an egg quality detection model to obtain an egg quality detection result output by the egg quality detection model, wherein the egg quality detection model is constructed on the basis of a YOLO model and combines a residual main network and an attention mechanism. According to the egg quality detection method, the pretreated egg image is input into the pre-trained egg quality detection model, the characteristics in the egg image are extracted by utilizing the egg quality detection model, and the egg quality detection result is obtained by automatic identification and detection, so that the egg quality detection efficiency and accuracy are effectively improved, and the labor cost is remarkably reduced.

Inventors

FANG CHENG
WU ZHENLONG
MA CHUANG
YANG JIKANG
Chao Xiaohuan
LUO QINGBIN
ZHANG XIQUAN
ZHANG TIEMIN

Assignees

华南农业大学

Dates

Publication Date: 20260512
Application Date: 20251212

Claims (10)

1. An egg quality detection method, which is characterized by comprising the following steps: acquiring an egg image of a target area; Preprocessing the egg image to obtain preprocessed egg image data characteristics; inputting the egg image data characteristics into an egg quality detection model to obtain an egg quality detection result output by the egg quality detection model; The egg quality detection model is constructed based on a YOLO model and combines a residual backbone network and an attention mechanism.
2. The method of claim 1, wherein the target area comprises a robot inspection area and a conveyor belt detection area, and wherein the acquiring an image of an egg of the target area comprises: based on a path planning algorithm, controlling a robot to carry out inspection in a poultry house according to a planned path, acquiring an inspection video through a camera carried by the robot, and extracting an image frame in the inspection video to serve as an egg image of an inspection area of the robot; And shooting and obtaining an egg image of the detection area of the conveyor belt through a camera at the tail end of the conveyor belt.
3. The method of claim 2, wherein the inputting the egg image data features into an egg quality detection model to obtain an egg quality detection result output by the egg quality detection model comprises: inputting the image data characteristics of the eggs in the inspection area of the continuous robots into an egg quality detection model; The chicken coop range is defined and the egg position is determined by identifying a chicken coop fence and a boundary frame of the egg in the image data characteristics of the egg; Detecting and obtaining an egg quality detection result of each egg in the range of the chicken coop; and outputting an egg quality detection result of the robot inspection area according to the change trend of the number of the fences.
4. The method of claim 2, wherein the inputting the egg image data features into an egg quality detection model to obtain an egg quality detection result output by the egg quality detection model comprises: inputting the image data characteristics of the eggs in the detection areas of a plurality of continuous conveyor belts into an egg quality detection model; Determining the integrity of eggs in the image data characteristics of each egg; detecting and obtaining an egg quality detection result of the egg image data characteristics of each egg with the integrity degree larger than a preset threshold value; and outputting an egg quality detection result of the detection area of the conveyor belt.
5. The method of claim 1, further comprising, after said inputting the image data characteristic of the egg into an egg quality detection model to obtain an egg quality detection result output by the egg quality detection model: uploading each egg image of the target area and a corresponding egg quality detection result to a cloud; Each egg image is marked with detection information, and the detection information comprises detection time, image numbers and detection object numbers.
6. The method of claim 1, wherein the egg quality detection model is trained based on the steps of: acquiring a plurality of historical egg images, and acquiring egg quality detection results corresponding to the historical egg images; preprocessing each historical egg image to obtain preprocessed historical egg image data characteristics; taking each historical egg image data characteristic as a sample, taking an egg quality detection result corresponding to each historical egg image data characteristic as a sample label corresponding to the sample, and constructing a training data set; and pre-training the egg quality detection model by using the training data set.
7. An egg quality inspection device, the device comprising: the image acquisition module is used for acquiring an egg image of the target area; The image processing module is used for preprocessing the egg image to obtain preprocessed egg image data characteristics; the image detection module is used for inputting the image data characteristics of the eggs into an egg quality detection model so as to obtain egg quality detection results output by the egg quality detection model; The egg quality detection model is constructed based on a YOLO model and combines a residual backbone network and an attention mechanism.
8. An electronic device comprising a memory storing a computer program and a processor implementing the method of any of claims 1 to 6 when the computer program is executed by the processor.
9. A computer readable storage medium storing a computer program, characterized in that the computer program, when executed by a processor, implements the method of any one of claims 1 to 6.
10. A computer program product comprising a computer program, characterized in that the computer program, when executed by a processor, implements the method of any one of claims 1 to 6.

Description

Egg quality detection method and related equipment Technical Field The application relates to the technical field of image detection, in particular to an egg quality detection method and related equipment. Background Currently, eggs are becoming increasingly interesting as important agricultural and food materials in terms of quality. In the related art, the recording of the egg yield and quality of the flock in the cage chicken farm is mainly carried out by a poultry house inspector in the chicken farm every day, and the egg yield and quality of all the cages are observed and recorded manually. However, in practical application, the conventional manual inspection method is heavy in process, so that the workload is high, and part of eggs are easy to be recorded in a missing mode. In summary, the technical problems in the related art are to be improved. Disclosure of Invention The embodiment of the application provides an egg quality detection method and related equipment, which can effectively improve the efficiency and accuracy of egg quality detection and remarkably reduce the labor cost. In one aspect, an embodiment of the present application provides a method for detecting egg quality, the method including the steps of: acquiring an egg image of a target area; Preprocessing the egg image to obtain preprocessed egg image data characteristics; inputting the egg image data characteristics into an egg quality detection model to obtain an egg quality detection result output by the egg quality detection model; The egg quality detection model is constructed based on a YOLO model and combines a residual backbone network and an attention mechanism. Optionally, the target area includes a robot inspection area and a conveyor belt detection area, and the acquiring the egg image of the target area includes: based on a path planning algorithm, controlling a robot to carry out inspection in a poultry house according to a planned path, acquiring an inspection video through a camera carried by the robot, and extracting an image frame in the inspection video to serve as an egg image of an inspection area of the robot; And shooting and obtaining an egg image of the detection area of the conveyor belt through a camera at the tail end of the conveyor belt. Optionally, the inputting the image data features of the eggs to an egg quality detection model to obtain an egg quality detection result output by the egg quality detection model includes: inputting the image data characteristics of the eggs in the inspection area of the continuous robots into an egg quality detection model; The chicken coop range is defined and the egg position is determined by identifying a chicken coop fence and a boundary frame of the egg in the image data characteristics of the egg; Detecting and obtaining an egg quality detection result of each egg in the range of the chicken coop; and outputting an egg quality detection result of the robot inspection area according to the change trend of the number of the fences. Optionally, the inputting the image data features of the eggs to an egg quality detection model to obtain an egg quality detection result output by the egg quality detection model includes: inputting the image data characteristics of the eggs in the detection areas of a plurality of continuous conveyor belts into an egg quality detection model; Determining the integrity of eggs in the image data characteristics of each egg; detecting and obtaining an egg quality detection result of the egg image data characteristics of each egg with the integrity degree larger than a preset threshold value; and outputting an egg quality detection result of the detection area of the conveyor belt. Optionally, after the inputting the image data features of the eggs to the egg quality detection model to obtain the egg quality detection result output by the egg quality detection model, the method further includes: uploading each egg image of the target area and a corresponding egg quality detection result to a cloud; Each egg image is marked with detection information, and the detection information comprises detection time, image numbers and detection object numbers. Optionally, the egg quality detection model is trained based on the following steps: acquiring a plurality of historical egg images, and acquiring egg quality detection results corresponding to the historical egg images; preprocessing each historical egg image to obtain preprocessed historical egg image data characteristics; taking each historical egg image data characteristic as a sample, taking an egg quality detection result corresponding to each historical egg image data characteristic as a sample label corresponding to the sample, and constructing a training data set; and pre-training the egg quality detection model by using the training data set. In another aspect, an embodiment of the present application provides an egg quality detection apparatus, including: the image acquisition module i