Pill Counting via Convex Contour Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing object counting systems are often time-consuming and prone to errors, especially when counting objects of varying sizes, shapes, or colors, as they often require manual effort or are limited to specific object recognition patterns, making them costly and less effective for diverse applications.
Innovation Solution
A digital image analysis apparatus comprising a light source, digital camera, textured surface, processing component, and display, which detects regular convex contours within images to count objects without relying on geometric profiles or training sets, allowing for accurate counting of objects like pills or tablets of different shapes and colors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If manual counting is used, then flexibility in handling diverse objects is maintained, but time consumption and error rates increase
Solution Approach 1:
The patent replaces manual mechanical counting with an automated digital imaging and image processing system. The system captures images of objects and uses computer vision algorithms to automatically count and identify objects, eliminating the need for manual intervention while maintaining the ability to handle diverse object types through pattern recognition
Solution Approach 2:
The system creates digital copies (images) of the physical objects and performs counting operations on these copies rather than manipulating the physical objects themselves. This allows for automated analysis while preserving the original objects and enabling rapid processing of multiple objects
2Measurement precision
If object-specific pattern recognition systems are used, then counting accuracy for specific objects improves, but system cost and complexity increase
Solution Approach 1:
The patent implements a universal image processing system that can count and identify multiple types of objects using the same hardware and software platform. The system uses general-purpose pattern recognition algorithms that can be applied to various object types without requiring separate specialized systems for each object category
Solution Approach 2:
The system achieves adaptability to different objects by changing processing parameters and thresholds in the image analysis algorithm rather than changing the physical system. By adjusting parameters such as contrast thresholds, shape tolerance ranges, and size filters, the same system can accurately count different types of objects with varying characteristics
3Measurement precision
If extensive training of object recognition patterns is performed, then recognition accuracy improves, but processing time and computational resources increase
Solution Approach 1:
The system performs preliminary processing of the image including enhancement, noise reduction, and feature extraction before the main counting operation. By preparing the image data in advance and identifying key features early in the process, the system reduces the computational burden of subsequent analysis and achieves faster overall processing
Solution Approach 2:
The image processing is divided into multiple sequential stages: image capture, preprocessing, feature detection, pattern recognition, and counting. Each stage processes only the necessary information for that specific task, avoiding redundant computations and enabling parallel processing where applicable, thus reducing total processing time
Data Source
AI summary
Systems and methods are described for counting objects by analyzing a digital image. The system or apparatus may include a light source, a digital camera, a textured surface disposed between the light source and the visible light camera, a processing component configured to produce a count of the objects, and a display configured to show the count of the objects. The method may include capturing an electronic image of the objects, detecting a plurality of edges within the image, identifying a plurality of concave sections based on the edges, identifying a regular convex contours based on the edges and the concave sections, determining whether the image is suitable for counting the objects based on the regular convex contours, and determining a count of the objects based the regular convex contours and the determination of whether the image is suitable.