Hatching Egg Fertilization Sorting via Near-Infrared Spectral Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for sorting unfertilized hatching eggs before incubation are inefficient, leading to resource waste and contamination, as they either fail to accurately detect unfertilized eggs before incubation or only detect after incubation has begun, and lack a subsequent sorting process.
Innovation Solution
A method and device using spectral feature fusion and a hatching egg classification neural network model to preprocess and classify fertilization information of hatching eggs, incorporating preprocessing steps like standard normal variable transformation and linear filtering, and a device with sensors and conveyor belts for sorting eggs based on time sequence features and fertilization information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual egg candling is used to remove infertile eggs, then the sorting process is simple to implement, but the detection speed is slow and labor costs are high
Solution Approach 1:
The patent replaces the manual mechanical egg candling process with an automated optical detection system using a camera and image processing algorithms. The system captures images of eggs and automatically analyzes them to distinguish fertilized from infertile eggs, eliminating the need for manual visual inspection while maintaining operational simplicity through automated decision-making.
Solution Approach 2:
The system enables self-service sorting by automatically detecting and classifying eggs without human intervention. The image processing system independently analyzes egg characteristics, makes classification decisions, and controls the sorting mechanism, allowing the system to serve itself in the detection and sorting task.
2Measurement precision
If computer vision technology is used to detect infertile eggs after incubation, then the detection accuracy is improved, but the detection time is too late causing resource waste
Solution Approach 1:
The patent applies preliminary action by detecting and identifying infertile eggs during the pre-incubation stage before the incubation process begins. The system analyzes eggs at an early stage when fertilization status can be determined, allowing timely removal of infertile eggs before they occupy incubation resources or undergo deterioration during incubation.
Solution Approach 2:
The system skips the traditional incubation period by performing detection before incubation starts. This rushing through of the detection process to an earlier time point enables identification and removal of infertile eggs without waiting for incubation to progress, thereby avoiding the time loss and resource waste associated with post-incubation detection.
3Difficulty of detecting and measuring
If X-ray device is used to detect embryo activity, then the detection capability is enhanced, but the detection timing is still too late when embryo development is required
Solution Approach 1:
The patent uses optical imaging and image processing to perform preliminary detection of egg fertilization status before incubation begins. This preliminary action allows the system to identify infertile eggs at an early stage, eliminating the need to wait for embryo development to detect fertility status, thereby avoiding the time loss that would occur with post-incubation detection methods like X-ray.
4Loss of time
If fiber optic spectroscopy technology is used to detect infertile eggs before incubation, then the detection timing is appropriate, but the data processing is insufficient reducing model accuracy
Solution Approach 1:
The patent extracts and selects key feature wavelengths from the spectral data through a wavelength selection algorithm. Instead of using all spectral data points, the system identifies and extracts the most informative wavelength features that are most discriminative for distinguishing fertilized from infertile eggs, thereby improving model accuracy while maintaining appropriate detection timing.
Solution Approach 2:
The system extracts and selects key feature wavelengths from the spectral data through a wavelength selection algorithm. Instead of using all spectral data points, the system identifies and extracts the most informative wavelength features that are most discriminative for distinguishing fertilized from infertile eggs, thereby improving model accuracy while maintaining appropriate detection timing.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Improves detection precision and efficiency by accurately identifying fertilized and unfertilized eggs before incubation, reducing waste and contamination, and automating the sorting process, thereby reducing personnel costs and optimizing production line efficiency.
Implementation Method 1
a near-infrared spectrometer, a halogen lamp light source and a display screen, the near-infrared spectrometer is configured to receive spectrum information transmitted by the sensor module to generate spectrum data
Data Source
AI summary
The present invention discloses a method and a device for non-destructive sorting of fertilization information of hatching eggs before incubation. The method comprises the following steps: obtaining a first time sequence feature of hatching eggs and collecting spectrum data of the hatching eggs according to the first time sequence feature; preprocessing the spectrum data; selecting the preprocessed spectrum data to obtain a feature band; constructing a hatching egg classification neural network model and training the hatching egg classification neural network model; classifying the hatching egg information according to the feature band by using the trained hatching egg classification neural network model to obtain fertilization information; and obtaining a second time sequence feature of the hatching eggs, and sorting the hatching eggs. The present invention improves the precision of detecting the fertilization information of hatching eggs.