Tofu Inspection and Sorting for Continuous Production Lines
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current tofu production systems rely heavily on human experience and visual inspection, leading to a high manual workload and limited productivity, with inadequate automation in conveyance and handling of defective products, and require a compact production system.
Innovation Solution
A tofu production system comprising a production device, conveyance devices, an inspection device, and a sorting and removing device that uses image capture and deep learning to identify defective products and automate their removal, reducing manual labor and improving production capacity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If human inspection is used to determine defective products, then flexibility in adjusting determination criteria is improved, but manual workload increases
Solution Approach 1:
The inspection system performs self-learning and self-adjustment of determination criteria through machine learning algorithms. The system automatically adapts to different product types and defect patterns without requiring manual intervention, thereby reducing manual workload while maintaining flexibility in adjusting determination criteria.
Solution Approach 2:
The patent replaces manual human inspection with an automated image recognition system using deep learning. This substitution eliminates the need for human operators to visually inspect products, significantly reducing manual workload while maintaining the ability to adjust determination criteria through software configuration.
2Productivity
If automated inspection is introduced to reduce manual workload, then productivity is improved, but system complexity increases
Solution Approach 1:
The inspection system is designed with multi-functionality to handle various product types and defect patterns using a single unified platform. The deep learning model can be trained on different datasets and applied to multiple inspection scenarios, reducing the need for multiple specialized systems and thereby managing complexity while improving productivity.
Solution Approach 2:
The patent introduces an intermediary layer of machine learning algorithms that bridge the gap between raw image data and inspection decisions. This intermediary processing layer simplifies the overall system architecture by automatically extracting features and making determinations, reducing the complexity of direct human-machine interaction while enhancing productivity.
3Ease of operation
If conventional conveyance systems are used, then ease of operation is maintained, but production capacity is limited
Solution Approach 1:
The conveyance system is designed to be dynamic and adaptable, with adjustable speeds and configurations that can be optimized for different production volumes and product types. This dynamic capability allows the system to maintain ease of operation while significantly improving production capacity through automated control and integration with the inspection system.
4Adaptability or versatility
If manual handling of defective products is used, then flexibility in disposal is improved, but time consumption increases
Solution Approach 1:
The patent replaces manual handling of defective products with an automated sorting and removal system. This mechanical substitution eliminates the time-consuming manual process of identifying and removing defective items, while the system maintains flexibility in disposal methods through programmable sorting mechanisms that can direct different types of defects to different destinations.
Data Source
AI summary
A tofu production system includes: a production device configured to continuously produce tofu; a conveyance device configured to arrange the tofu produced by the production device according to a predetermined rule corresponding to the tofu and convey the tofu; a tofu inspection device configured to inspect the tofu on the conveyance device; and a sorting and removing device configured to sort or remove a defective product in the tofu being conveyed by the conveyance device based on an inspection result of the tofu inspection device.