scholarly journals Part Identification System for a Flexible Feeder using Proximity Sensors

2015 ◽  
Vol 14 (3-4) ◽  
Author(s):  
K Sadesh ◽  
PV Mohram ◽  
S Udhayakumar
Keyword(s):  
Low Cost ◽  
Identification System ◽  
Arbitrary Orientation ◽  
Vision Systems ◽  
Flexible Part

<p>A part feeder intakes identical parts of arbitrary orientation and provides output in uniform orientation. A flexible feeder is capable of handling parts of several sizes. The two important modules of a flexible part feeder are (i) identification of part (ii) adjustments to accommodate the incoming part. This paper aims at first module i.e. developing a low cost part identification system using two proximity sensors and thereby eliminating the use of costlier vision systems. The proposed part identification system using two capacitive type proximity sensors was effective in identifying the size of incoming parts and the efficiency of the system was around 84.6%.</p>

Real time fall detection using infrared cameras and reflective tapes under day/night luminance

2021 ◽  
pp. 1-16
Author(s):  
E. Ramanujam ◽  
S. Padmavathi
Keyword(s):  
Low Cost ◽  
Age Groups ◽  
Orientation Angle ◽  
Fall Detection ◽  
Living Environment ◽  
Vision Systems ◽  
Ir Camera ◽  
Cluttered Environments ◽  
Infrared Cameras ◽  
Time Fall

Falls are the leading cause of injuries and death in elderly individuals who live alone at home. The core service of assistive living technology is to monitor elders’ activities through wearable devices, ambient sensors, and vision systems. Vision systems are among the best solutions, as their implementation and maintenance costs are the lowest. However, current vision systems are limited in their ability to handle cluttered environments, occlusion, illumination changes throughout the day, and monitoring without illumination. To overcome these issues, this paper proposes a 24/7 monitoring system for elders that uses retroreflective tape fabricated as part of conventional clothing, monitored through low-cost infrared (IR) cameras fixed in the living environment. IR camera records video even when there are changes in illumination or zero luminance. For classification among clutter and occlusion, the tape is considered as a blob instead of a human silhouette; the orientation angle, fitted through ellipse modeling, of the blob in each frame allows classification that detects falls without pretrained data. System performance was tested using subjects in various age groups and “fall” or “non-fall” were detected with 99.01% accuracy.

Fingerprint Identification System Based on DSP

2012 ◽  
Vol 468-471 ◽  
pp. 920-923
Author(s):  
Ya Ping Bao ◽  
Li Liu ◽  
Yuan Wang ◽  
Qian Song
Keyword(s):  
Software Design ◽  
High Speed ◽  
High Efficiency ◽  
Low Cost ◽  
Fingerprint Identification ◽  
Identification System ◽  
System Operation ◽  
Operation Speed ◽  
High Speed Data Acquisition ◽  
High Speed Data

This paper introduced a fast fingerprint identification system based on TMS320VC5416 DSP chip and MBF200 solidity fingerprint sensor. It precipitates fingerprint identification device developing into the direction of miniaturization, embedded and automatic.It recommends fingerprint identification system hardware and software design and the main system processing flow, aim at fingerprint identification arithmetic, the influence of system operation speed is being researched at the same time. High-speed data acquisition system is been built in order to achieve a DSP fingerprint identification system with high efficiency and low cost.

A dsPIC Microcontroller Based System Identification of Positive Output Buck Boost Converter for Module Mismatch Application

2015 ◽  
Vol 785 ◽  
pp. 106-110
Author(s):  
M.N.M. Hussain ◽  
Ahmad Maliki Omar ◽  
Intan Rahayu Ibrahim ◽  
Kamarulazhar Daud
Keyword(s):  
Direct Method ◽  
Low Cost ◽  
Digital Signal ◽  
Solar Irradiation ◽  
Identification System ◽  
Pv System ◽  
Partial Shading ◽  
Single Output ◽  
Pv Module ◽  
Power Point

An identification system of multiple-input single-output (MISO) model is developed in controlling dsPIC microcontroller of positive output buck-boost (POBB) converters for module mismatch condition of photovoltaic (PV) system. In particular, the possibility of the scheme is to resolve the mismatch losses from the PV module either during shading or mismatch module occurrences. The MPPT algorithm is simplified by identification approach of indirect incorporated with a simple incremental direct method to form a combined direct and indirect (CoDId) algorithms. Irregular consumption of solar irradiation on a PV module shall step-up or step down the voltage regarding to the desired DC output voltage of POBB converter. This optimized algorithm will ensure that the PV module to kept at maximum power point (MPP), preventing power loss during module mismatch incident in PV module especially during partial shading condition. The simulation and laboratory results for PV module of polycrystalline Mitsubishi PV-AE125MF5N indicate that the proposed model and development of PV system architecture performs well, while the efficiency up to 97.7% at critical of low solar irradiance level. The controlling signal is based on low-cost embedded microcontroller of dsPIC30F Digital Signal Control (DSC).

scholarly journals Practical Implementation of an Indoor Robot Localization and Identification System using an Array of Anchor Nodes

2020 ◽  
Vol 16 (1) ◽  
pp. 1-8 ◽  
Author(s):  
Israa AL-Forati ◽  
Abdulmuttalib Rashid
Keyword(s):  
Search Algorithm ◽  
Low Cost ◽  
Binary Search ◽  
Practical Implementation ◽  
Identification System ◽  
Robot Localization ◽  
Light Emitting ◽  
The Matrix ◽  
Anchor Nodes ◽  
Binary Search Algorithm

This paper proposes a low-cost Light Emitting Diodes (LED) system with a novel arrangement that allows an indoor multi-robot localization. The proposed system uses only a matrix of low-cost LED installed uniformly on the ground of an environment and low-cost Light Dependent Resistor (LDR), each equipped on bottom of the robot for detection. The matrix of LEDs which are driven by a modified binary search algorithm are used as active beacons. The robot localizes itself based on the signals it receives from a group of neighbor LEDs. The minimum bounded circle algorithm is used to draw a virtual circle from the information collected from the neighbor LEDs and the center of this circle represents the robot’s location. The propose system is practically implemented on an environment with (16*16) matrix of LEDs. The experimental results show good performance in the localization process.

Drought Stress Detection Using Low-Cost Computer Vision Systems and Machine Learning Techniques

2020 ◽  
Vol 22 (3) ◽  
pp. 27-29 ◽  
Author(s):  
Paula Ramos-Giraldo ◽  
Chris Reberg-Horton ◽  
Anna M. Locke ◽  
Steven Mirsky ◽  
Edgar Lobaton
Keyword(s):  
Machine Learning ◽  
Computer Vision ◽  
Drought Stress ◽  
Low Cost ◽  
Machine Learning Techniques ◽  
Stress Detection ◽  
Vision Systems ◽  
Learning Techniques ◽  
Computer Vision Systems

Bluetooth-Based Room Localization Research Based on NB and SVM Approach

2014 ◽  
Vol 926-930 ◽  
pp. 2458-2461
Author(s):  
Qing Chen ◽  
Peng Ming Wang
Keyword(s):  
Home Environment ◽  
System Development ◽  
Low Cost ◽  
Identification System ◽  
Bayes Classifier ◽  
Integrated Environment ◽  
Vector Machines ◽  
Mobile Phone Network ◽  
Hybrid Classification ◽  
Localization Systems

It has been general recognized that the application of localization technology in home environment are beneficial to the development of health monitoring and mobile identification system development. As a kind of highly efficient sensor with obvious advantages such as low cost, the Bluetooth device has been widely used in our daily life. Research is carried out in an integrated environment based on mobile phone network signal measurement and Bluetooth link measurements in developing home localization systems. This paper presented a hybrid classification method, based on the combination of Bayesian network and supported vector machines, to support the development of Bluetooth-based room localization system. The proposed method mainly considers the dependency between features and non-linear overlapping of features between rooms. The results show that the prediction accuracy has been improved greatly in comparison to the traditional Naive Bayes classifier and the hidden Markov model used in previous studies.

Sign in / Sign up

Export Citation Format

Share Document