Real-time on-line unconstrained handwriting recognition using statistical methods

2002 ◽  
Author(s):  
K.S. Nathan ◽  
H.S.M. Beigi ◽  
Jayashree Subrahmonia ◽  
G.J. Clary ◽  
H. Maruyama
Keyword(s):  
Real Time ◽  
Statistical Methods ◽  
Handwriting Recognition ◽  
On Line

scholarly journals Monitoring of the process of Flash-Butt Welding

2013 ◽  
Vol 18 (1) ◽  
pp. 31-38 ◽  
Author(s):  
Yevgenia Chvertko ◽  
Mykola Shevchenko ◽  
Andriy Pirumov
Keyword(s):  
Neural Networks ◽  
Real Time ◽  
Statistical Methods ◽  
Welded Joints ◽  
Welding Parameter ◽  
Monitoring Systems ◽  
Butt Welding ◽  
On Line ◽  
Welding Processes

Statistical methods of analysis are currently widely used to develop control and monitoring systems for different welding processes. These methods allow to obtain information about the process including effect of all factors on its results, which is often difficult to evaluate due to the complexity of the process. The authors made efforts to apply these methods to develop the system for monitoring the parameters of flash-butt welding in real-time mode. The paper gives brief information about the features of flash-butt welding of reinforcement bars and some basic limitation of this process application. The main reasons of formation of defects in welded joints are given as well as analysis of possibility of application of monitoring systems for their determination. The on-line monitoring system based on neural networks was developed for evaluation of process deviations. This system is believed to be adequate for determination of process violations resulting in disturbances of welding parameter and can be used for prediction of possible defects in the welded joints.

ON-LINE RUN-ON CHARACTER RECOGNIZER: DESIGN AND PERFORMANCE

1991 ◽  
Vol 05 (01n02) ◽  
pp. 123-137 ◽  
Author(s):  
T. FUJISAKI ◽  
T. E. CHEFALAS ◽  
J. KIM ◽  
C. C. TAPPERT ◽  
C. G. WOLF
Keyword(s):  
Real Time ◽  
Handwriting Recognition ◽  
Computer Interface ◽  
On Line ◽  
Recognition Systems ◽  
Linguistic Constraints ◽  
And Performance ◽  
Character Sequence

A recognize-then-segment recognizer of unconstrained handprinting uses a unified tablet-display to provide a paper-like computer interface. Whereas most handwriting recognition systems segment and then recognize, this one recognizes and then finds the best segmentation. It classifies strokes, generates character hypotheses, and verifies hypotheses to estimate the optimal character sequence for each word of run-on handwritten characters. Linguistic constraints can limit the choices. The system is implemented on an IBM workstation, accepts run-on characters written on a tablet, and performs recognition in real time.

Adaptive Signal Analysis and Interpretation for Real-time Intelligent Patient Monitoring

1994 ◽  
Vol 33 (01) ◽  
pp. 60-63 ◽  
Author(s):  
E. J. Manders ◽  
D. P. Lindstrom ◽  
B. M. Dawant
Keyword(s):  
Computer Architecture ◽  
Real Time ◽  
Data Abstraction ◽  
Monitoring Strategy ◽  
Intelligent Monitoring ◽  
Patient Status ◽  
On Line ◽  
Main Components ◽  
And Control ◽  
Adaptive Signal

Abstract:On-line intelligent monitoring, diagnosis, and control of dynamic systems such as patients in intensive care units necessitates the context-dependent acquisition, processing, analysis, and interpretation of large amounts of possibly noisy and incomplete data. The dynamic nature of the process also requires a continuous evaluation and adaptation of the monitoring strategy to respond to changes both in the monitored patient and in the monitoring equipment. Moreover, real-time constraints may imply data losses, the importance of which has to be minimized. This paper presents a computer architecture designed to accomplish these tasks. Its main components are a model and a data abstraction module. The model provides the system with a monitoring context related to the patient status. The data abstraction module relies on that information to adapt the monitoring strategy and provide the model with the necessary information. This paper focuses on the data abstraction module and its interaction with the model.

Nitrogen Removal in a Real-Time Controlled Sequencing Batch Membrane Bioreactor

2010 ◽  
Vol 5 (3) ◽  
Author(s):  
Cheng-Nan Chang ◽  
Li-Ling Lee ◽  
Han-Hsien Huang ◽  
Ying-Chih Chiu
Keyword(s):  
Real Time ◽  
Nitrogen Removal ◽  
Membrane Bioreactor ◽  
Synthetic Wastewater ◽  
Real Time Control ◽  
Time Control ◽  
Cake Layer ◽  
On Line ◽  
Solid Liquid ◽  
Solid Liquid Separation

The performance of a real-time controlled Sequencing Batch Membrane Bioreactor (SBMBR) for removing organic matter and nitrogen from synthetic wastewater has been investigated in this study under two specific ammonia loadings of 0.0086 and 0.0045g NH4+-N gVSS−1 day−1. Laboratory results indicate that both COD and DOC removal are greater than 97.5% (w/w) but the major benefit of using membrane for solid-liquid separation is that the effluent can be decanted through the membrane while aeration is continued during the draw stage. With a continued aeration, the sludge cake layer is prevented from forming thus alleviating the membrane clogging problem in addition to significant nitrification activities observed in the draw stage. With adequate aeration in the oxic stage, the nitrogen removal efficiency exceeding 99% can be achieved with the SBMBR system. Furthermore, the SBMBR system has also been used to study the occurrence of ammonia valley and nitrate knee that can be used for real-time control of the biological process. Under appropriate ammonia loading rates, applicable ammonia valley and nitrate knee are detected. The real-time control of the SBMBR can be performed based on on-line ORP and pH measurements.

Modifications of rainfall runoff and decision finding models for on-line simulation in real time control

1999 ◽  
Vol 39 (9) ◽  
pp. 201-207
Author(s):  
Andreas Cassar ◽  
Hans-Reinhard Verworn
Keyword(s):  
Real Time ◽  
Urban Drainage ◽  
Rain Event ◽  
Rainfall Runoff ◽  
Real Time Control ◽  
Design Storm ◽  
Time Control ◽  
Timing Data ◽  
On Line ◽  
Rainfall Runoff Model

Most of the existing rainfall runoff models for urban drainage systems have been designed for off-line calculations. With a design storm or a historical rain event and the model system the rainfall runoff processes are simulated, the faster the better. Since very recently, hydrodynamic models have been considered to be much too slow for real time applications. However, with the computing power of today - and even more so of tomorrow - very complex and detailed models may be run on-line and in real time. While the algorithms basically remain the same as for off-line simulations, problems concerning timing, data management and inter process communication have to be identified and solved. This paper describes the upgrading of the existing hydrodynamic rainfall runoff model HYSTEM/EXTRAN and the decision finding model INTL for real time performance, their implementation on a network of UNIX stations and the experiences from running them within an urban drainage real time control project. The main focus is not on what the models do but how they are put into action and made to run smoothly embedded in all the processes necessary in operational real time control.

scholarly journals Learning Feedforward Control Using Multiagent Control Approach for Motion Control Systems

Energies ◽  
2021 ◽  
Vol 14 (2) ◽  
pp. 420
Author(s):  
Phong B. Dao
Keyword(s):  
Real Time ◽  
Feedforward Control ◽  
Design Method ◽  
Mechatronic Systems ◽  
Control Approach ◽  
Runtime Environment ◽  
Multiagent Control ◽  
On Line ◽  
Promising Solution ◽  
Function Approximator

Multiagent control system (MACS) has become a promising solution for solving complex control problems. Using the advantages of MACS-based design approaches, a novel solution for advanced control of mechatronic systems has been developed in this paper. The study has aimed at integrating learning control into MACS. Specifically, learning feedforward control (LFFC) is implemented as a pattern for incorporation in MACS. The major novelty of this work is that the feedback control part is realized in a real-time periodic MACS, while the LFFC algorithm is done on-line, asynchronously, and in a separate non-real-time aperiodic MACS. As a result, a MACS-based LFFC design method has been developed. A second-order B-spline neural network (BSN) is used as a function approximator for LFFC whose input-output mapping can be adapted during control and is intended to become equal to the inverse model of the plant. To provide real-time features for the MACS-based LFFC system, the open robot control software (OROCOS) has been employed as development and runtime environment. A case study using a simulated linear motor in the presence of nonlinear cogging and friction force as well as mass variations is used to illustrate the proposed method. A MACS-based LFFC system has been designed and implemented for the simulated plant. The system consists of a setpoint generator, a feedback controller, and a time-index LFFC that can learn on-line. Simulation results have demonstrated the applicability of the design method.

scholarly journals DOLARS, a Distributed On-Line Activity Recognition System by Means of Heterogeneous Sensors in Real-Life Deployments—A Case Study in the Smart Lab of The University of Almería

Sensors ◽  
2021 ◽  
Vol 21 (2) ◽  
pp. 405
Author(s):  
Marcos Lupión ◽  
Javier Medina-Quero ◽  
Juan F. Sanjuan ◽  
Pilar M. Ortigosa
Keyword(s):  
Real Time ◽  
Activity Recognition ◽  
Real Life ◽  
Recognition System ◽  
Machine Learning Algorithms ◽  
Sensor Data ◽  
Heterogeneous Sensors ◽  
On Line ◽  
The University

Activity Recognition (AR) is an active research topic focused on detecting human actions and behaviours in smart environments. In this work, we present the on-line activity recognition platform DOLARS (Distributed On-line Activity Recognition System) where data from heterogeneous sensors are evaluated in real time, including binary, wearable and location sensors. Different descriptors and metrics from the heterogeneous sensor data are integrated in a common feature vector whose extraction is developed by a sliding window approach under real-time conditions. DOLARS provides a distributed architecture where: (i) stages for processing data in AR are deployed in distributed nodes, (ii) temporal cache modules compute metrics which aggregate sensor data for computing feature vectors in an efficient way; (iii) publish-subscribe models are integrated both to spread data from sensors and orchestrate the nodes (communication and replication) for computing AR and (iv) machine learning algorithms are used to classify and recognize the activities. A successful case study of daily activities recognition developed in the Smart Lab of The University of Almería (UAL) is presented in this paper. Results present an encouraging performance in recognition of sequences of activities and show the need for distributed architectures to achieve real time recognition.

Sign in / Sign up

Export Citation Format

Share Document