Fast and Accurate Convolution Neural Network for Detecting Manufacturing Data

Authors:
Publisher website
Google Scholar
Abstract
This paper introduces a Clustering with Particle for Object Detection (CPOD) for object detection in smart factories. CPOD builds on regional based methods to identify smart object data using outlier detection, clustering, particle swarm optimization (PSO), and deep convolutional networks. The process starts by removing noise and errors from the images database by Local Outlier Factor algorithm. Next, the algorithm studies different correlations from the set of images in the database. PSO is used to optimize the different parameters including, the number of neighbours of LOF, the number of clusters of k-means, the number of epochs, and the error learning rate for Fast-RCNN. The inference process benefits from the knowledge provided by training. Instead of considering a complex single model of the whole images database, we consider a simple homogeneous model. Our results reveal the superiority of CPOD against baseline object detection solutions, in terms of runtime and accuracy.

This publication has 29 references indexed in Scilit: