Intelligent Foreign Particle Inspection Machine for Injection Liquid Examination Based on Modified Pulse-Coupled Neural Networks
Open Access
Abstract
A biologically inspired spiking neural network model, called pulse-coupled neural networks (PCNN), has been applied in an automatic inspection machine to detect visible foreign particles intermingled in glucose or sodium chloride injection liquids. Proper mechanisms and improved spin/stop techniques are proposed to avoid the appearance of air bubbles, which increases the algorithms’ complexity. Modified PCNN is adopted to segment the difference images, judging the existence of foreign particles according to the continuity and smoothness properties of their moving traces. Preliminarily experimental results indicate that the inspection machine can detect the visible foreign particles effectively and the detection speed, accuracy and correct detection rate also satisfying the needs of medicine preparation.Keywords
This publication has 14 references indexed in Scilit:
- Motion layer extraction in the presence of occlusion using graph cutsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2005
- Efficient Graph-Based Image SegmentationInternational Journal of Computer Vision, 2004
- A technique for making holograms easily and for measuring simultaneously the behaviour of particles of different sizes and/or shapesMeasurement Science and Technology, 2004
- Pattern recognition using pulse-coupled neural networks and discrete Fourier transformsNeurocomputing, 2003
- An adaptive neuro-fuzzy system for automatic image segmentation and edge detectionIEEE Transactions on Fuzzy Systems, 2002
- Normalized cuts and image segmentationIEEE Transactions on Pattern Analysis and Machine Intelligence, 2000
- Adaptive integrated image segmentation and object recognitionIEEE Transactions on Systems, Man and Cybernetics, Part C (Applications and Reviews), 2000
- PCNN models and applicationsIEEE Transactions on Neural Networks, 1999
- Adaptive image segmentation using a genetic algorithmIEEE Transactions on Systems, Man, and Cybernetics, 1995
- Feature Linking via Synchronization among Distributed Assemblies: Simulations of Results from Cat Visual CortexNeural Computation, 1990