Multi-Spectral Waste Sorting for Plastic Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automated waste sorting technologies, particularly those using optical and image analysis in the visible spectrum, face challenges in accurately distinguishing between materials like plastics, transparent objects, and nested items, leading to low recognition performance and contamination issues, especially in achieving high purity levels required for recycling.
Innovation Solution
A system utilizing multiple light sources, including visible light and ultraviolet light, to enhance image contrast and discrimination, combined with infrared classification, for improved material recognition and sorting, facilitating better identification and separation of waste materials.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If optical sorting machines using light spectrometry in the infrared spectrum are used, then material identification is possible through database comparison, but the performance in distinguishing between various materials is not completely satisfactory, especially for plastics material
Solution Approach 1:
The patent divides the detection process into multiple spectral segments by using separate detection devices for different wavelength ranges (visible light, near-infrared, short-wave infrared). Each detection device captures specific material characteristics in its optimized spectral band, and the results are combined to achieve comprehensive material identification, thereby improving reliability for plastics sorting.
Solution Approach 2:
The patent extends the detection from a single spectral dimension (traditional infrared) to multiple spectral dimensions by incorporating visible light and near-infrared detection. This multi-dimensional spectral analysis provides additional distinguishing features for materials, particularly plastics, enabling more accurate differentiation between material types.
2Extent of automation
If image analysis in the visible spectrum using white light and cameras is used, then object recognition is possible through shape and appearance, but the performance is below expectations with recognition levels of around 85%, and transparent objects are little visible
Solution Approach 1:
The patent segments the imaging process into multiple spectral components by using separate camera systems for visible light and near-infrared wavelengths. Each camera captures different material properties - visible light provides color and shape information while near-infrared reveals transparency and material composition, allowing the system to overcome the limitations of single-spectrum imaging.
Solution Approach 2:
The patent creates a composite detection system that combines multiple spectral detection channels (visible, near-infrared, short-wave infrared) into a unified material identification process. This composite approach integrates information from different spectral domains to form a comprehensive material signature, significantly improving recognition accuracy for transparent and complex plastic objects.
3Ease of manufacture
If single spectrum detection is used, then the system is simple to implement, but there are few contrasts and distinguishing factors for identification, particularly for transparent objects and nested objects
Solution Approach 1:
The patent segments the detection function across multiple spectral bands, with each detection device optimized for its specific wavelength range. This segmentation allows each component to remain relatively simple while the collective system gains enhanced discriminatory power through the combination of spectral information from visible, near-infrared, and short-wave infrared channels.
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
The multi-light source approach significantly improves the accuracy and purity of sorted waste, achieving recognition performance up to 95% and higher, surpassing conventional methods by providing additional visual information and fluorescence characteristics for better material differentiation.
Implementation Method 1
at least one second light source, with a nature distinct from the first light source, allowing the appearance of additional visual information on said at least one image acquired... said at least one second light source emits an ultraviolet light allowing the appearance of fluorescence in the visible spectrum
Implementation Method 2
the objects are irradiated by a light spectrum, generally in the infrared lengths, sometimes supplemented by complementary analyses, and a sensor records the reflected spectrum
Data Source
AI summary
Disclosed is a system and a method for classifying articles in a flow of articles to be separated, the flow of articles to be separated being installed on a conveying device, including: —an image acquisition member installed so as to be able to acquire at least one image of a portion of the flow of articles to be separated; —a first overhanging light source, which emits in the visible spectrum and illuminates the portion of the flow of articles to be separated, the at least one image of which is acquired by the image acquisition member; a classification member capable of classifying the articles of the portion of the flow of articles to be separated according to the acquired image; and at least one second light source, of a different nature than the first light source, allowing additional visual information to appear on the acquired image.