Performance Study of the Application of Artificial Neural Networks to the Completion and Prediction of Data Retrieved by Underwater Sensors

Self Cite

This paper presents an intelligent surveillance platform based on the usage of large numbers of inexpensive sensors designed and developed inside the European Eureka Celtic project HuSIMS. With the aim of maximizing the number of deployable units while keeping monetary and resource/bandwidth costs at a minimum, the surveillance platform is based on the usage of inexpensive visual sensors which apply efficient motion detection and tracking algorithms to transform the video signal in a set of motion parameters. In order to automate the analysis of the myriad of data streams generated by the visual sensors, the platform's control center includes an alarm detection engine which comprises three components applying three different Artificial Intelligence strategies in parallel. These strategies are generic, domain-independent approaches which are able to operate in several domains (traffic surveillance, vandalism prevention, perimeter security, etc.). The architecture is completed with a versatile communication network which facilitates data collection from the visual sensors and alarm and video stream distribution towards the emergency teams. The resulting surveillance system is extremely suitable for its deployment in metropolitan areas, smart cities, and large facilities, mainly because cheap visual sensors and autonomous alarm detection facilitate dense sensor network deployments for wide and detailed coverage.

Section: Monitoring and Control Systemmentioning

confidence: 99%

An Intelligent Surveillance Platform for Large Metropolitan Areas with Dense Sensor Deployment

Fernández

Calavia

Baladrón

et al. 2013

Self Cite

“…Various studies on the recovery of lost data have mostly been focused on the failure tolerance of the wireless sensor networks [19,20]. The concept of large-scale neuron sensor networks or a carbon nanotubes (CNT)-based artificial neural system (ANS) [11,12,21] or the use of artificial neural networks in management of strain data in a sensor network [22,23] may be used in estimation of strain data without recovery of abnormal sensors.…”

Section: Introductionmentioning

confidence: 99%

A Practical Data Recovery Technique for Long-Term Strain Monitoring of Mega Columns during Construction

Choi

Kwon

Kim

et al. 2013

A practical data recovery method is proposed for the strain data lost during the safety monitoring of mega columns. The analytical relations among the measured strains are derived to recover the data lost due to unexpected errors in long-term measurement during construction. The proposed technique is applied to recovery of axial strain data of a mega column in an irregular building structure during construction. The axial strain monitoring using the wireless strain sensing system was carried out for one year and five months between 23 July 2010 and 22 February 2012. During the long-term strain sensing, three different types of measurement errors occurred. Using the recovery technique, the strain data that could not be measured at different intervals in the measurement were successfully recovered. It is confirmed that the problems that may occur during long-term wireless strain sensing of mega columns during construction could be resolved through the proposed recovery method.

“…Monitoring systems are usually employed to acquire and store the most useful variables of a process. In the literature, these systems have been proposed to monitor atmospheric phenomena such as air pollution [7], livestock farms such as aviary production systems [8], or environments including rivers [9], oceans [10], and ecological systems [11]. In the field of agriculture, monitoring systems have been proposed for indoor locations such as greenhouses [12–14], and for outdoor agricultural processes including wine harvesting [15], olive growing [16], or any generic process [17].…”

Section: Introductionmentioning

confidence: 99%

Temperature and Relative Humidity Estimation and Prediction in the Tobacco Drying Process Using Artificial Neural Networks

Martínez-Martínez

Baladrón

Gómez-Gil

et al. 2012

Self Cite

This paper presents a system based on an Artificial Neural Network (ANN) for estimating and predicting environmental variables related to tobacco drying processes. This system has been validated with temperature and relative humidity data obtained from a real tobacco dryer with a Wireless Sensor Network (WSN). A fitting ANN was used to estimate temperature and relative humidity in different locations inside the tobacco dryer and to predict them with different time horizons. An error under 2% can be achieved when estimating temperature as a function of temperature and relative humidity in other locations. Moreover, an error around 1.5 times lower than that obtained with an interpolation method can be achieved when predicting the temperature inside the tobacco mass as a function of its present and past values with time horizons over 150 minutes. These results show that the tobacco drying process can be improved taking into account the predicted future value of the monitored variables and the estimated actual value of other variables using a fitting ANN as proposed.