Automated Medicine Sorting via Image Discrimination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing medicine sorting devices cannot automatically sort medicines such as tablets or capsules that are not packaged in containers, as they are designed to handle ampoules or vials, leaving a gap in automated sorting solutions for these types of medications.
Innovation Solution
A medicine sorting device equipped with an imaging part to capture images of medicines, a discrimination part to identify medicine types, and a sorting part to categorize and store them by type, utilizing features extraction and comparison with a medicine database for accurate sorting, even if medicines are not pre-registered.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual sorting by pharmacist or physician is used, then medicines can be sorted by type, but time consumption and labor effort increase significantly
Solution Approach 1:
The patent replaces the manual mechanical sorting process with an automated imaging and recognition system. The imaging part captures images of medicines, the discrimination part identifies medicine types through image analysis, and the sorting part automatically categorizes medicines by type, eliminating manual labor while maintaining high sorting accuracy.
Solution Approach 2:
The system enables medicines to be automatically identified and sorted through their own visual characteristics. The imaging part captures images of the medicines themselves, and the discrimination part extracts features from these images to identify medicine types without requiring external labeling or manual intervention.
2Productivity
If returned medicines are discarded without sorting, then time and effort for sorting are saved, but medication reuse opportunities are lost and resources are wasted
Solution Approach 1:
The system creates visual copies (images) of the medicines through the imaging part. These images are then analyzed by the discrimination part to identify medicine types, enabling automatic sorting decisions based on visual information rather than physical handling or complex identification procedures.
Solution Approach 2:
The discrimination part changes the state of medicine identification from manual visual inspection to automated image feature extraction. By extracting features such as shape, size, color, and packaging characteristics from images, the system transforms unidentifiable mixed medicines into categorized sorted medicines based on multiple visual parameters.
3Extent of automation
If automated sorting is implemented for ampoules or vials, then sorting efficiency improves, but medicines such as tablets or capsules that are not packaged in containers cannot be sorted
Solution Approach 1:
The imaging part is designed to capture images of medicines in various forms including tablets, capsules, and packaged medicines. The discrimination part extracts features from these diverse images to identify different medicine types, enabling the system to handle multiple medicine formats universally rather than being limited to specific container types.
Solution Approach 2:
The system dynamically adapts its recognition approach based on the medicine type detected in the image. For packaged medicines, it may recognize packaging characteristics, while for unpackaged medicines like tablets or capsules, it focuses on the medicine本身的 visual features such as shape, size, and color, making the sorting process flexible and adaptable to different medicine forms.
Data Source
AI summary
A medicine sorting device includes: a discrimination part for discriminating a type of a medicine based on an image captured by a first camera; and a conveyance/sorting unit for sorting, by each type, a plurality of types of medicines accommodated in a mixed state in a first accommodating part based on a discrimination result of the discrimination part, and storing the medicines in a second accommodating part.