Intelligent injection liquid particle inspection machine based on two-dimensional Tsallis Entropy with modified pulse-coupled neural networks

  • Authors:

  • Yaonan Wang;Ji Ge;Hui Zhang;Bowen Zhou

  • Affiliations:

  • College of Electrical and Information Engineering, Hunan University, Changsha 410082, China;College of Electrical and Information Engineering, Hunan University, Changsha 410082, China;College of Electrical and Information Engineering, Hunan University, Changsha 410082, China;College of Electrical and Information Engineering, Hunan University, Changsha 410082, China

  • Venue:
  • Engineering Applications of Artificial Intelligence
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

The Automatic Liquid Particle Inspection Machine (AIM) using 2-D Tsallis Entropy with modified pulse-coupled neural networks (PCNN) is used in order to detect visible foreign particles within injection fluids. According to the motion of the particles in liquid, appropriate mechanisms are utilized which guarantees that the inspection machine will follow detection procedures: ''Rotation, Abruptly Braking, Video Tracking'' to extract tiny objects from complicated sequential images. In order to reduce the influence derived from air bubbles, improved spin/stop techniques are applied. The external capture mode of CCD cameras is used to avoid the possibility of omitting certain particles by trivial displacement. 2-D Tsallis Entropy with modified PCNN is applied in order to segment the difference images, and then to judge the existence of foreign particles according to the continuity and smoothness of their traces. Preliminary experimental results (125ml 0.9% sodium chloride solution and 10% glucose as the samples) indicate that the inspection machine, which is superior to proficient inspectors, can detect the visible foreign particles effectively and that this detection speed and accuracy, as well as the correct detection rate can also facilitate the medicinal construction.