Fast object recognition and classification is highly important when handling operations with robots. This article shows the design and implementation of an invariant recognition machine vision system to compute a descriptive vector called the Boundary Object Function (BOF) using the FuzzyARTMAP (FAM) Neural Network. The object recognition machine is integrated in the Zybo Z7-20 module that includes reconfigurable FPGA hardware and a RISC processor. Object encoding, description and prediction is carried out rapidly compared to the processing time devoted to video capture at the camera’s frame rate. Benefiting from parallel computing, we calculated the object’s centroid and boundary points while acquiring the progressive image frame; all that was done with the intention of readying it for neural processing. The remaining time was devoted to recognising the object, this caused low latency (1.47 ms). Our test-bed also included TCP/IP communication to send/receive part location for grasping operations with an industrial robot to evaluate the approach. Results demonstrate that the hardware integration of the video sensor, image processing, descriptor generator, and the ANN classifier for cognitive decision on a single chip can increase the speed and performance of intelligent robots designed for smart manufacturing.

No abstract

The process of recognizing manufacturing parts in real time requires fast, accurate, small, and low-power-consumption sensors. Here, we describe a method to extract descriptors from several objects observed from a wide range of angles in a three-dimensional space. These descriptors define the dataset, which allows for the training and further validation of a convolutional neural network. The classification is implemented in reconfigurable hardware in an embedded system with an RGB sensor and the processing unit. The system achieved an accuracy of 96.67% and a speed 2.25× faster than the results reported for state-of-the-art solutions. Our proposal is 655 times faster than implementation on a PC. The presented embedded system meets the criteria of real-time video processing and it is suitable as an enhancement for the hand of a robotic arm in an intelligent manufacturing cell.

The article presents a method for obtaining the contour of an object in real time from non-binarized images for recognition purpose. The contour information is integrated into a descriptive vector named BOF used by a FuzzyARTMAP Artificial Neural Network (ANN) model to learn the object and then recognize it later. In this way, it is possible to obtain a learning process regarding the location and recognition of parts; to communicate to a robot arm the position and orientation information of an object for assembly purposes. Other method to obtain contour using binarized images, is compared with the described method in this paper in order to implement and test both in a Field Programmable Gate Array (FPGA) architecture. Since an ANN can be implemented more efficiently in a parallel structure such as FPGA architecture can supply, it is desirable to implement an efficient algorithm for obtaining the object contour in the same way.

No abstract