Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Deep learning-based egg group image segmentation fertilization information detection device and method

A deep learning and image segmentation technology, applied in neural learning methods, measuring devices, material analysis by optical means, etc., can solve the problems of unsuitable large-scale rapid detection automatic production, expensive spectral equipment, complicated operation, etc., to save incubation Space, improve detection efficiency, simple operation effect

Pending Publication Date: 2022-05-27
HENAN UNIV OF SCI & TECH
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] However, spectral equipment is too expensive for low-cost agricultural production, and has high requirements for the detection environment, and the objects are all single eggs, which limits the promotion of these methods; although the above methods can detect the hatching of early eggs, the operation Complicated and costly, it is not suitable for large-scale rapid detection and to meet the needs of current automated production
At present, the hatching quality detection method of breeding eggs is single egg detection, which is low in efficiency and high in cost, and there is little research on the group detection method of hatching quality

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Deep learning-based egg group image segmentation fertilization information detection device and method
  • Deep learning-based egg group image segmentation fertilization information detection device and method
  • Deep learning-based egg group image segmentation fertilization information detection device and method

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach

[0058] like Image 6 As shown, the fertilization information detection device for group egg image segmentation based on deep learning: including a transmission platform, a platform bracket 2, a special bilateral conveyor belt 8, a driving wheel, a driven wheel, a tensioning device 9, a support cylinder, a servo motor 6, and a photoelectric sensor 13 , the egg tray positioning cylinder 12, the industrial depth camera 10, the camera sliding bracket 11, the egg tray device 1 and the corresponding egg camera device 14, the depth camera 10 is arranged directly above the egg tray device 1, the said The depth camera 10 is vertically connected with the camera sliding bracket 11; the photoelectric sensor 13, the egg tray positioning cylinder 12 and the computer 16 are connected with the programmable controller 15; The moving wheel rotates to drive the conveyor to run, and the tensioning device 9 takes effect to prevent the conveyor belt from loosening.

[0059] The photoelectric senso...

Embodiment 1

[0088] Fertilization information detection method based on deep learning group egg image segmentation:

[0089] First, the images of the 3d, 4d, 5d, 6d, 7d, and 8d days of hatching of group eggs 17 were obtained through the built detection equipment, and median filtering was performed on the images collected for several days to remove the salt and pepper noise in the images. The color and texture feature classification of the picture is realized by obtaining the region of interest (ROI) by combining image segmentation and deep learning. In order to improve the speed and effect of dividing the picture, the transmission image of the egg 17 is preliminarily cropped, and the cropped image is divided into 30 pieces. ROI area, obtain the ROI center coordinates of a single egg 17.

[0090] Create a mask with the same size as the segmentation, draw a white solid circle with the size of ROI on the mask, use the mask to perform the "AND" operation, and retain the ROI area through the co...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides a deep learning-based egg group image segmentation fertilization information detection device and method, and the method comprises the following steps: firstly collecting a deep egg group transmission image, segmenting the egg group transmission image into a single target region through a self-designed image segmentation method, building a deep hatching egg fertilization information detection model, and carrying out the recognition of the egg group transmission image. Image texture features are extracted through an optimized MobileNetV1 neural network model, fertilization information in the hatching egg hatching process is detected online, and nondestructive detection of the hatching egg hatching process of the group is achieved; the device is simple to operate, can detect a plurality of hatching eggs at one time, is high in production efficiency, can realize nondestructive detection of hatching egg activity, is suitable for forming large-scale rapid detection and meets the requirements of current automatic production; the establishment method of the model is simple to operate, the model can be used for realizing the nondestructive fertilization detection at the early hatching stage of hatching eggs, the production cost is reduced, and the detection efficiency is improved.

Description

technical field [0001] The invention belongs to the field of detection of fertilized eggs, in particular to a device and method for detecting fertilization information based on deep learning based on group egg image segmentation. Background technique [0002] It takes about 21 days for the eggs to hatch into chicks. In order to ensure the normal development of the chicken embryos, the incubation environment needs to maintain a certain temperature and humidity. Failure to remove infertile eggs or dead embryo eggs in time during the incubation process not only takes up space in the incubator and wastes incubation resources, but also harmful microorganisms such as mold and bacteria multiply rapidly in the incubator, and even infect other normally hatched eggs, resulting in hatching The rate is reduced, resulting in huge economic losses. Therefore, it is of great significance to promptly remove infertile eggs in the early hatching of breeding eggs. [0003] At present, the mos...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G01N21/84G01N21/01G01N33/08G06V20/68G06V10/25G06V10/26G06V10/56G06V10/54G06V10/764G06V10/82G06K9/62G06N3/04G06N3/08
CPCG01N21/84G01N21/01G01N33/085G06N3/08G01N2021/845G01N2021/0112G06N3/045G06F18/24Y02A40/70
Inventor 张伏王顺青曹炜桦王新月崔夏华禹煌张朝臣滕帅邱玉博张亚坤王甲甲付三玲
Owner HENAN UNIV OF SCI & TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products