Results 1 - 10
of
12
Limited receptive area neural classifier for recognition of swallowing sounds using short-time Fourier transform
"... Abstract — In this paper we propose a sound recognition technique based on the limited receptive area (LIRA) neural classifier and short-time Fourier transform (STFT). LIRA neural classifier was developed as a multipurpose image recognition system. Previous tests of LIRA demonstrated good results in ..."
Abstract
-
Cited by 6 (4 self)
- Add to MetaCart
(Show Context)
Abstract — In this paper we propose a sound recognition technique based on the limited receptive area (LIRA) neural classifier and short-time Fourier transform (STFT). LIRA neural classifier was developed as a multipurpose image recognition system. Previous tests of LIRA demonstrated good results in different image recognition tasks including: handwritten digit recognition, face recognition, metal surface texture recognition, and micro work piece shape recognition. We propose a sound recognition technique where spectrograms of sound instances serve as inputs of the LIRA neural classifier. The methodology was tested in recognition of swallowing sounds. Swallowing sound recognition may be employed in systems for automated swallowing assessment and diagnosis of swallowing disorders. The experimental results suggest high efficiency and reliability of the proposed approach. I.
Recognition of Swallowing Sounds Using Time- Frequency Decomposition and Limited Receptive Area Neural Classifier
"... Abstract In this paper we propose a novel swallowing sound recognition technique based on the limited receptive area (LIRA) neural classifier and timefrequency decomposition. Time-frequency decomposition methods commonly used in sound recognition increase dimensionality of the signal and require ste ..."
Abstract
-
Cited by 2 (0 self)
- Add to MetaCart
(Show Context)
Abstract In this paper we propose a novel swallowing sound recognition technique based on the limited receptive area (LIRA) neural classifier and timefrequency decomposition. Time-frequency decomposition methods commonly used in sound recognition increase dimensionality of the signal and require steps of feature selection and extraction. Quite often feature selection is based on a set of empirically chosen statistics, making the pattern recognition dependent on the intuition and skills of the investigator. A limited set of extracted features is then presented to a classifier. The proposed method avoids the steps of feature selection and extraction by delegating them to a limited receptive area neural (LIRA) classifier. LIRA neural classifier utilizes the increase in dimensionality of the signal to create a large number of random features in the time-frequency domain that assure a good description of the signal without prior assumptions of the signal properties. Features that do not provide useful information for separation of classes do not obtain significant weights during classifier training. The proposed methodology was tested on the task of recognition of swallowing sounds with two
Hierarchically Clustered Adaptive Quantization CMAC and Its Learning Convergence
"... Abstract—The cerebellar model articulation controller (CMAC) neural network (NN) is a well-established computational model of the human cerebellum. Nevertheless, there are two major drawbacks associated with the uniform quantization scheme of the CMAC network. They are the following: 1) a constant o ..."
Abstract
- Add to MetaCart
Abstract—The cerebellar model articulation controller (CMAC) neural network (NN) is a well-established computational model of the human cerebellum. Nevertheless, there are two major drawbacks associated with the uniform quantization scheme of the CMAC network. They are the following: 1) a constant output resolution associated with the entire input space and 2) the generalization-accuracy dilemma. Moreover, the size of the CMAC network is an exponential function of the number of inputs. Depending on the characteristics of the training data, only a small percentage of the entire set of CMAC memory cells is utilized. Therefore, the efficient utilization of the CMAC memory is a crucial issue. One approach is to quantize the input space nonuniformly. For existing nonuniformly quantized CMAC systems, there is a tradeoff between memory efficiency and computational complexity. Inspired by the underlying organizational mechanism of the human brain, this paper presents a novel CMAC architecture named hierarchically clustered adaptive quantization CMAC (HCAQ-CMAC). HCAQ-CMAC employs hierarchical clustering for the nonuniform quantization of the input space to identify significant input segments and subsequently allocating more memory cells to these regions. The stability of the HCAQ-CMAC network is theoretically guaranteed by the proof of its learning convergence. The performance of the proposed network is subsequently benchmarked against the original CMAC network, as well as two other existing CMAC variants on two real-life applications, namely, automated control of car maneuver and modeling of the human blood glucose dynamics. The experimental results have demonstrated that the HCAQ-CMAC network offers an efficient memory allocation scheme and improves the generalization and accuracy of the network output to achieve better or comparable performances with smaller memory usages. Index Terms—Cerebellar model articulation controller (CMAC), hierarchical clustering, hierarchically clustered adaptive quantization CMAC (HCAQ-CMAC), learning convergence, nonuniform quantization.
Development and Implementation of the LIRA Neural Classifier
"... Abstract—Neural network is a tool in the solution of control problems. Thanks to their capacity for learning it is possible to train neural networks to recognize different patterns. The patterns involve diverse conditions of operation under which the system must be trained to be able to make the dec ..."
Abstract
- Add to MetaCart
(Show Context)
Abstract—Neural network is a tool in the solution of control problems. Thanks to their capacity for learning it is possible to train neural networks to recognize different patterns. The patterns involve diverse conditions of operation under which the system must be trained to be able to make the decisions to control the system or process. One of the methods of neural network simulation is digital design. We begin from a neuron design, and then we simulate the neural network that can be applied to the control process. The schematic design of logic circuits allows investigation of the circuit behavior and verification that the circuit fulfills the desired goals. Index Terms—LIRA neural classifier, logic circuits, neural networks, neuron.
Proceedings of the 29th Annual International Conference of the IEEE EMBS
"... Abstract — A significant part of daily energy expenditure may be attributed to non-exercise activity thermogenesis and exercise activity thermogenesis. Automatic recognition of postural allocations such as standing or sitting can be used in behavioral modification programs aimed at minimizing static ..."
Abstract
- Add to MetaCart
(Show Context)
Abstract — A significant part of daily energy expenditure may be attributed to non-exercise activity thermogenesis and exercise activity thermogenesis. Automatic recognition of postural allocations such as standing or sitting can be used in behavioral modification programs aimed at minimizing static postures. In this paper we propose a shoe-based device and related pattern recognition methodology for recognition of postural allocations. Inexpensive technology allows implementation of this methodology as a part of footwear. The experimental results suggest high efficiency and reliability of the proposed approach. I.
swallowing sounds using continuous wavelet transform
"... Abstract — In this paper we propose a sound recognition technique based on the limited receptive area (LIRA) neural classifier and continuous wavelet transform (CWT). LIRA neural classifier was developed as a multipurpose image recognition system. Previous tests of LIRA demonstrated good results in ..."
Abstract
- Add to MetaCart
(Show Context)
Abstract — In this paper we propose a sound recognition technique based on the limited receptive area (LIRA) neural classifier and continuous wavelet transform (CWT). LIRA neural classifier was developed as a multipurpose image recognition system. Previous tests of LIRA demonstrated good results in different image recognition tasks including: handwritten digit recognition, face recognition, metal surface texture recognition, and micro work piece shape recognition. We propose a sound recognition technique where scalograms of sound instances serve as inputs of the LIRA neural classifier. The methodology was tested in recognition of swallowing sounds. Swallowing sound recognition may be employed in systems for automated swallowing assessment and diagnosis of swallowing disorders. The experimental results suggest high efficiency and reliability of the proposed approach. I.
The biologically inspired Hierarchical Temporal Memory Model for Farsi Handwritten Digit and Letter Recognition
"... It is herein proposed a handwritten digit recognition system which biologically inspired of the large-scale structure of the mammalian neocortex. Hierarchical Temporal Memory (HTM) is a memory-prediction network model that takes advantage of the Bayesian belief propagation and revision techniques. I ..."
Abstract
- Add to MetaCart
It is herein proposed a handwritten digit recognition system which biologically inspired of the large-scale structure of the mammalian neocortex. Hierarchical Temporal Memory (HTM) is a memory-prediction network model that takes advantage of the Bayesian belief propagation and revision techniques. In this article a study has been conducted to train a HTM network to recognize handwritten digits and letters taken from the well-known Hoda dataset for Farsi handwritten digit. Results presented in this paper show good performance and generalization capacity of the proposed network for a real-world big dataset. Keywords Handwritten digit recognition; hierarchical temporal memory (HTM); Hoda handwritten digits dataset. 1.
Non-profit academic project, developed under the open access initiative A Scatter Search Algorithm for Solving a Bilevel Optimization Model for Determining Highway Tolls
"... How to cite Complete issue More information about this article Journal's homepage in redalyc.org Scientific Information System ..."
Abstract
- Add to MetaCart
How to cite Complete issue More information about this article Journal's homepage in redalyc.org Scientific Information System
Flat Facet Parabolic Solar Concentrator With Support Cell for One and More Mirrors
"... Abstract:- Parabolic dish solar concentrators are very expensive devices with a cost of up to a half of the total cost of a solar power station. The specific technology of parabolic mirror manufacturing makes the decrease of the cost very problematic. Even in mass production the cost of a parabolic ..."
Abstract
- Add to MetaCart
(Show Context)
Abstract:- Parabolic dish solar concentrators are very expensive devices with a cost of up to a half of the total cost of a solar power station. The specific technology of parabolic mirror manufacturing makes the decrease of the cost very problematic. Even in mass production the cost of a parabolic mirror is estimated as $500 per square meter. There is another way to make a parabolic concentrator by approximating a parabolic surface with large number of small flat mirrors. We created a small prototype of this type of solar concentrator which contains 24 flat mirrors in the form of equilateral triangles with side length of 50mm. This prototype has special nuts to adjust the positions of the nodes in the points of connections of the apexes. These nuts make it possible to approximate the parabolic shape in an easy and inexpensive way.
Support Frame for Micro Facet Solar Concentrator
"... Abstract:- The low cost micro facet solar concentrator is proposed. Large number of small flat mirrors is situated at parabolic surface to approximate large parabolic mirror. Low cost commercial flat mirrors can be used for manufacturing of such concentrator. The problems of production of micro mirr ..."
Abstract
- Add to MetaCart
(Show Context)
Abstract:- The low cost micro facet solar concentrator is proposed. Large number of small flat mirrors is situated at parabolic surface to approximate large parabolic mirror. Low cost commercial flat mirrors can be used for manufacturing of such concentrator. The problems of production of micro mirrors, support components and automatic assembly of concentrator are discussed. Rough estimations show that the cost of the concentrator should be approximately $ 55 per square meter of concentrator surface. Key-Words:- solar concentrator, flat micro mirror, automatic assembly 1
