Meal Box Defect Detection Using Image Acquirer and Server Model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for detecting defects in meal boxes are prone to subjective errors, lack uniformity in judgment criteria, and have low accuracy and efficiency, often relying on manual or semi-automatic optical detection that cannot cover all detection criteria and is not expandable.
Innovation Solution
A method and device that utilize an image acquirer on a user terminal to capture images of meal boxes, which are then transmitted to a server for defect recognition using a trained defect detection model, enabling real-time, automatic, and visually-based defect detection, including normalization of pixel data to reduce errors and improve detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual or semi-automatic optical detection is used for meal box quality inspection, then the detection process can be performed with simple equipment, but the detection accuracy and efficiency are low and subjective errors occur
Solution Approach 1:
The patent replaces manual mechanical inspection with an automated optical detection system using image acquirers (cameras) and deep learning algorithms. The system captures images of meal boxes and uses trained defect detection models to automatically identify defects, eliminating subjective human errors while maintaining relatively simple hardware composition.
Solution Approach 2:
The patent creates a virtual copy of the meal box through image acquisition, then performs defect detection on the digital image rather than the physical object. This allows multiple detection criteria to be applied simultaneously to the same image without additional physical sensors, improving both accuracy and efficiency.
2Productivity
If manual detection methods are used for meal box quality inspection, then the detection criteria can be flexibly adjusted, but the detection efficiency and productivity are low
Solution Approach 1:
The patent implements continuous automated detection by capturing images of meal boxes as they move through the production line and processing them in real-time through the defect detection model. This eliminates idle time between inspections and maintains continuous detection operation, significantly improving productivity compared to intermittent manual inspection.
Solution Approach 2:
The detection system performs self-service by automatically capturing images, processing them through the trained defect detection model, and generating detection results without requiring manual intervention for each inspection. The system autonomously completes the entire detection workflow, freeing operators from repetitive manual inspection tasks.
3Adaptability or versatility
If semi-automatic detection systems are used for meal box inspection, then some automation is achieved, but the system lacks expandability and cannot cover all detection criteria
Solution Approach 1:
The patent implements a universal defect detection model that can detect multiple types of defects (foreign objects, packaging defects, food quality issues) using a single integrated system. The deep learning model is trained on diverse defect types and can identify various defects across different meal box categories, providing multi-functional detection capability without requiring separate specialized systems for each defect type.
Solution Approach 2:
The patent employs a dynamic and adaptable detection system where the defect detection model can be retrained and updated with new defect types and criteria. The system's detection capabilities can evolve and expand over time by incorporating new training data and adjusting detection parameters, allowing it to adapt to changing product requirements and emerging defect types without hardware modifications.
Data Source
AI summary
Embodiments of the present disclosure provide a method and a device for detecting defect of a meal box, a server, a device, and a storage medium. The method includes: receiving a detection request including an image of the meal box sent by a user terminal, the image of the meal box being obtained by an image acquirer of the user terminal; and performing defect recognition based on the image of the meal box and a defect detection model in response to the detection request.