Special Issue "Neural Networks and Sensors"

Quicklinks

A special issue of Algorithms (ISSN 1999-4893).

Deadline for manuscript submissions: closed (31 July 2009)

Special Issue Editor

Guest Editor
Dr. Michael W. Retsky

Harvard School of Public Health, 677 Huntington Avenue, Boston, MA 02115, USA
University College London, Glower Street, London WC1E 6BT, UK
Website | E-Mail
Phone: +1-203-675-0017
Interests: cancer research; electron beam technology

Keywords

  • neural networks
  • artificial intelligence
  • artificial neural networks
  • sensors

Related Special Issue

Published Papers (8 papers)

View options order results:
result details:
Displaying articles 1-8
Export citation of selected articles as:

Research

Jump to: Review

Open AccessArticle Classification of Echolocation Calls from 14 Species of Bat by Support Vector Machines and Ensembles of Neural Networks
Algorithms 2009, 2(3), 907-924; doi:10.3390/a2030907
Received: 16 June 2009 / Accepted: 1 July 2009 / Published: 9 July 2009
Cited by 25 | PDF Full-text (83 KB) | HTML Full-text | XML Full-text
Abstract
Calls from 14 species of bat were classified to genus and species using discriminant function analysis (DFA), support vector machines (SVM) and ensembles of neural networks (ENN). Both SVMs and ENNs outperformed DFA for every species while ENNs (mean identification rate – 97%)
[...] Read more.
Calls from 14 species of bat were classified to genus and species using discriminant function analysis (DFA), support vector machines (SVM) and ensembles of neural networks (ENN). Both SVMs and ENNs outperformed DFA for every species while ENNs (mean identification rate – 97%) consistently outperformed SVMs (mean identification rate – 87%). Correct classification rates produced by the ENNs varied from 91% to 100%; calls from six species were correctly identified with 100% accuracy. Calls from the five species of Myotis, a genus whose species are considered difficult to distinguish acoustically, had correct identification rates that varied from 91 – 100%. Five parameters were most important for classifying calls correctly while seven others contributed little to classification performance. Full article
(This article belongs to the Special Issue Neural Networks and Sensors)
Open AccessArticle Neural Network Modeling to Predict Shelf Life of Greenhouse Lettuce
Algorithms 2009, 2(2), 623-637; doi:10.3390/a2020623
Received: 24 October 2008 / Revised: 7 March 2009 / Accepted: 25 March 2009 / Published: 3 April 2009
Cited by 2 | PDF Full-text (89 KB) | HTML Full-text | XML Full-text
Abstract
Greenhouse-grown butter lettuce (Lactuca sativa L.) can potentially be stored for 21 days at constant 0°C. When storage temperature was increased to 5°C or 10°C, shelf life was shortened to 14 or 10 days, respectively, in our previous observations. Also, commercial shelf life
[...] Read more.
Greenhouse-grown butter lettuce (Lactuca sativa L.) can potentially be stored for 21 days at constant 0°C. When storage temperature was increased to 5°C or 10°C, shelf life was shortened to 14 or 10 days, respectively, in our previous observations. Also, commercial shelf life of 7 to 10 days is common, due to postharvest temperature fluctuations. The objective of this study was to establish neural network (NN) models to predict the remaining shelf life (RSL) under fluctuating postharvest temperatures. A box of 12 - 24 lettuce heads constituted a sample unit. The end of the shelf life of each head was determined when it showed initial signs of decay or yellowing. Air temperatures inside a shipping box were recorded. Daily average temperatures in storage and averaged shelf life of each box were used as inputs, and the RSL was modeled as an output. An R2 of 0.57 could be observed when a simple NN structure was employed. Since the "future" (or remaining) storage temperatures were unavailable at the time of making a prediction, a second NN model was introduced to accommodate a range of future temperatures and associated shelf lives. Using such 2-stage NN models, an R2 of 0.61 could be achieved for predicting RSL. This study indicated that NN modeling has potential for cold chain quality control and shelf life prediction. Full article
(This article belongs to the Special Issue Neural Networks and Sensors)
Figures

Open AccessArticle A Sensor-Based Learning Algorithm for the Self-Organization of Robot Behavior
Algorithms 2009, 2(1), 398-409; doi:10.3390/a2010398
Received: 30 November 2008 / Accepted: 26 February 2009 / Published: 4 March 2009
Cited by 8 | PDF Full-text (2618 KB) | HTML Full-text | XML Full-text
Abstract
Ideally, sensory information forms the only source of information to a robot. We consider an algorithm for the self-organization of a controller. At short time scales the controller is merely reactive but the parameter dynamics and the acquisition of knowledge by an internal
[...] Read more.
Ideally, sensory information forms the only source of information to a robot. We consider an algorithm for the self-organization of a controller. At short time scales the controller is merely reactive but the parameter dynamics and the acquisition of knowledge by an internal model lead to seemingly purposeful behavior on longer time scales. As a paradigmatic example, we study the simulation of an underactuated snake-like robot. By interacting with the real physical system formed by the robotic hardware and the environment, the controller achieves a sensitive and body-specific actuation of the robot. Full article
(This article belongs to the Special Issue Neural Networks and Sensors)
Open AccessArticle Self-organization of Dynamic Distributed Computational Systems Applying Principles of Integrative Activity of Brain Neuronal Assemblies
Algorithms 2009, 2(1), 247-258; doi:10.3390/a2010247
Received: 28 November 2008 / Revised: 26 January 2009 / Accepted: 16 February 2009 / Published: 17 February 2009
Cited by 1 | PDF Full-text (336 KB) | HTML Full-text | XML Full-text
Abstract
This paper presents a method for self-organization of the distributed systems operating in a dynamic context. We propose the use of a simple biologically (cognitive neuroscience) inspired method for system configuration that allows allocating most of the computational load to off-line in order
[...] Read more.
This paper presents a method for self-organization of the distributed systems operating in a dynamic context. We propose the use of a simple biologically (cognitive neuroscience) inspired method for system configuration that allows allocating most of the computational load to off-line in order to improve the scalability property of the system. The method proposed has less computational burden at runtime than traditional system adaptation approaches. Full article
(This article belongs to the Special Issue Neural Networks and Sensors)
Open AccessArticle The Autonomous Stress Indicator for Remotely Monitoring Power System State and Watching for Potential Instability
Algorithms 2009, 2(1), 183-199; doi:10.3390/a2010183
Received: 9 October 2008 / Revised: 15 January 2009 / Accepted: 2 February 2009 / Published: 10 February 2009
PDF Full-text (722 KB) | HTML Full-text | XML Full-text
Abstract
The proposed Autonomous Stress Indicator (ASI) is a device that monitors the contents of the protection relays on a suspect weak power system bus and generates a performance level related to the degree of system performance degradation or instability. This gives the system
[...] Read more.
The proposed Autonomous Stress Indicator (ASI) is a device that monitors the contents of the protection relays on a suspect weak power system bus and generates a performance level related to the degree of system performance degradation or instability. This gives the system operators some time (minutes) to take corrective action. In a given operating area there would not likely be a need for an ASI on every bus. Note that the ASI does not trip any breakers; it is an INFORMATION ONLY device. An important feature is that the system operator can subsequently interrogate the ASI to determine the factor(s) that led to the performance level that has been initially annunciated, thereby leading to a course of action. This paper traces the development of the ASI which is an ongoing project. The ASI could be also described as a stress-alert device whose function is to alert the System Operator of a stressful condition at its location. The characteristics (or essential qualities) of this device are autonomy, selectivity, accuracy and intelligence. These will fulfill the requirements of the recommendation of the Canada –US Task Force in the August 2003 system collapse. Preliminary tests on the IEEE 39-bus model indicate that the concept has merit and development work is in progress. While the ASI can be applied to all power system operating conditions, its principal application is to the degraded state of the system where the System Operator must act to restore the system to the secure state before it migrates to a stage of collapse. The work of ASI actually begins with the Areas of Vulnerability and ends with the Predictive Module as described in detail in this paper. An application example of a degraded system using the IEEE 39-bus system is included. Full article
(This article belongs to the Special Issue Neural Networks and Sensors)

Review

Jump to: Research

Open AccessReview Advances in Artificial Neural Networks – Methodological Development and Application
Algorithms 2009, 2(3), 973-1007; doi:10.3390/algor2030973
Received: 1 July 2009 / Revised: 24 July 2009 / Accepted: 28 July 2009 / Published: 3 August 2009
Cited by 36 | PDF Full-text (622 KB) | HTML Full-text | XML Full-text
Abstract
Artificial neural networks as a major soft-computing technology have been extensively studied and applied during the last three decades. Research on backpropagation training algorithms for multilayer perceptron networks has spurred development of other neural network training algorithms for other networks such as radial
[...] Read more.
Artificial neural networks as a major soft-computing technology have been extensively studied and applied during the last three decades. Research on backpropagation training algorithms for multilayer perceptron networks has spurred development of other neural network training algorithms for other networks such as radial basis function, recurrent network, feedback network, and unsupervised Kohonen self-organizing network. These networks, especially the multilayer perceptron network with a backpropagation training algorithm, have gained recognition in research and applications in various scientific and engineering areas. In order to accelerate the training process and overcome data over-fitting, research has been conducted to improve the backpropagation algorithm. Further, artificial neural networks have been integrated with other advanced methods such as fuzzy logic and wavelet analysis, to enhance the ability of data interpretation and modeling and to avoid subjectivity in the operation of the training algorithm. In recent years, support vector machines have emerged as a set of high-performance supervised generalized linear classifiers in parallel with artificial neural networks. A review on development history of artificial neural networks is presented and the standard architectures and algorithms of artificial neural networks are described. Furthermore, advanced artificial neural networks will be introduced with support vector machines, and limitations of ANNs will be identified. The future of artificial neural network development in tandem with support vector machines will be discussed in conjunction with further applications to food science and engineering, soil and water relationship for crop management, and decision support for precision agriculture. Along with the network structures and training algorithms, the applications of artificial neural networks will be reviewed as well, especially in the fields of agricultural and biological engineering. Full article
(This article belongs to the Special Issue Neural Networks and Sensors)
Open AccessReview Actual Pathogen Detection: Sensors and Algorithms - a Review
Algorithms 2009, 2(1), 301-338; doi:10.3390/a2010301
Received: 10 December 2008 / Revised: 13 February 2009 / Accepted: 24 February 2009 / Published: 3 March 2009
Cited by 12 | PDF Full-text (304 KB) | HTML Full-text | XML Full-text
Abstract
Pathogens feed on fruits and vegetables causing great food losses or at least reduction of their shelf life. These pathogens can cause losses of the final product or in the farms were the products are grown, attacking leaves, stems and trees. This review
[...] Read more.
Pathogens feed on fruits and vegetables causing great food losses or at least reduction of their shelf life. These pathogens can cause losses of the final product or in the farms were the products are grown, attacking leaves, stems and trees. This review analyses disease detection sensors and algorithms for both the farm and postharvest management of fruit and vegetable quality. Mango, avocado, apple, tomato, potato, citrus and grapes were selected as the fruits and vegetables for study due to their world-wide consumption. Disease warning systems for predicting pathogens and insects on farms during fruit and vegetable production are commonly used for all the crops and are available where meteorological stations are present. It can be seen that these disease risk systems are being slowly replaced by remote sensing monitoring in developed countries. Satellite images have reduced their temporal resolution, but are expensive and must become cheaper for their use world-wide. In the last 30 years, a lot of research has been carried out in non-destructive sensors for food quality. Actually, non-destructive technology has been applied for sorting high quality fruit which is desired by the consumer. The sensors require algorithms to work properly; the most used being discriminant analysis and training neural networks. New algorithms will be required due to the high quantity of data acquired and its processing, and for disease warning strategies for disease detection. Full article
(This article belongs to the Special Issue Neural Networks and Sensors)
Open AccessReview Neural Network Analysis and Evaluation of the Fetal Heart Rate
Algorithms 2009, 2(1), 19-30; doi:10.3390/a2010019
Received: 21 November 2008 / Revised: 4 January 2009 / Accepted: 8 January 2009 / Published: 16 January 2009
Cited by 12 | PDF Full-text (344 KB) | HTML Full-text | XML Full-text
Abstract
The aim of the present study is to obtain a highly objective automatic fetal heart rate (FHR) diagnosis. The neural network software was composed of three layers with the back propagation, to which 8 FHR data, including sinusoidal FHR, were input and the
[...] Read more.
The aim of the present study is to obtain a highly objective automatic fetal heart rate (FHR) diagnosis. The neural network software was composed of three layers with the back propagation, to which 8 FHR data, including sinusoidal FHR, were input and the system was educated by the data of 20 cases with a known outcome. The output was the probability of a normal, intermediate, or pathologic outcome. The neural index studied prolonged monitoring. The neonatal states and the FHR score strongly correlated with the outcome probability. The neural index diagnosis was correct. The completed software was transferred to other computers, where the system function was correct. Full article
(This article belongs to the Special Issue Neural Networks and Sensors)

Planned Papers

The below list represents only planned manuscripts. Some of these manuscripts have not been received by the Editorial Office yet. Papers submitted to MDPI journals are subject to peer-review.


Journal Contact

MDPI AG
Algorithms Editorial Office
St. Alban-Anlage 66, 4052 Basel, Switzerland
algorithms@mdpi.com
Tel. +41 61 683 77 34
Fax: +41 61 302 89 18
Editorial Board
Contact Details Submit to Algorithms
Back to Top