A data processing apparatus may utilize an artificial neuron network configured to reduce dimensionality of input data using a sparse transformation configured using receptive field structure of network units. Output of the network may be analyzed for temporally persistency that is characterized by similarity matrix. Elements of the matrix may be incremented when present activity unit activity at a preceding frame. The similarity matrix may be partitioned based on a distance measure for a given element of the matrix and its closest neighbors. Stability of learning of temporally proximal patterns may be greatly improved as the similarity matrix is learned independently of the partitioning operation. Partitioning of the similarity matrix using the methodology of the disclosure may be performed online, e.g., contemporaneously with the encoding and/or similarity matrix construction, thereby enabling learning of new features in the input data.

Apparatus and methods for temporal proximity detection

In Indonesia, the demands of air tranportation for reaching destination increase rapidly. Based on the flight schedule in airports spreading in Indonesia, the airports have different flight demand rate so that it requires clustering. This research will use two methods for clustering : kohonen network and Fuzzy C Means (FCM).Kohonen network is the type neural network which uses unsupervised training.Kohonen network uses weight vectors for training while FCM uses degree of membership. Both kohonen network and FCM, inputs are represented by the number of departure and arrival of airline in one day. For kohonen network, we update weight matrices so that minimizing the sum of optimum euclidean distance. For FCM, we update degrees of membership so that minimizing the objective function value.From the simulations, we can cluster the airports based on the number of departure and arrival of airline.

/pdf/application-kohonen-network-and-fuzzy-c-means-for-clustering-3bpexlnefh.pdf

Application Kohonen Network and Fuzzy C Means for Clustering Airports Based on Frequency of Flight

Selecting a desired level of quality in a product is crucial—for instance, in dried chili peppers (Capsicum annuum L.). However, classifying dried chili peppers is a time-consuming, manual task. One of the main problems with sorting this way is the lack of product homogeneity. This paper presents the development of a classification system, based on artificial neural networks, for size and color recognition applied to the aforementioned peppers. The classification system uses 8-bit grayscale-image histograms to characterize the peppers. Three quality levels identified in a Mexican Official Norm (NMX-FF-107/1-SCFI-2014) for a specific type of chili (Guajillo) were used to show the proposed method. The effectiveness of the proposed classification system is demonstrated through experiments and is measured using the receiver-operating characteristic curve, calculating the area under the curve and obtaining an accuracy of 82.13. The results show an innovative, reliable and economical alternative, capable of sorting dried chili peppers; a system that aims to contribute to the solution of the problem of identification and classification in dehydrators y/o final customers.

A novel method for dried chili pepper classification using artificial intelligence

Over the last decades, the development of information technology in the field of computer systems and smart-devices has been increasing rapidly. Especially, due to the improvement of smart-devices, they have become a necessary ubiquitous device influencing our daily life. Nowadays, with the popularity of computer systems and smart-devices, the exposure to security risks has been on the rise throughout the world. Therefore, systems security (e.g. computer systems and smart-devices security) has become a significant concern. In order to alleviate security risks, cybersecurity researchers have concentrated more on presenting novel, high-qualified, and integrated security systems.

A novel host intrusion detection system using neural network

Face recognition is the process of identifying one or more people in images or videos. It is an important part of biometric, security & surveillance system, and image indexing systems. Various face recognition techniques have been proposed in literature such as: Eigen-faces, Feature based, Hidden Markov model and Neural network based techniques. The first three techniques mostly include a phase of feature extraction or preprocessing closely related to the type of image to recognize. On the other hand Neural network technique does not need specific data about the type of image, thus can be applied to any type of image and at the same time provides better accuracy. In this paper we made an effort to combine neural network technique with fuzzy logic. Our experimental result shows that combining the two provide better accuracy in comparison to other techniques mentioned above.

http://www.sersc.org/journals/IJSIP/vol7_no1/31.pdf

Face Recognition using Neuro-Fuzzy Inference System

Face Recognition has been identified as one of the attracting research areas and it has drawn the attention of many researchers due to its varying applications such as security systems, medical systems, entertainment, etc. Face recognition is the preferred mode of identification by humans: it is natural, robust and non-intrusive. A wide variety of systems requires reliable personal recognition schemes to either confirm or determine the identity of an individual requesting their services. The purpose of such schemes is to ensure that the rendered services are accessed only by a legitimate user and no one else. Examples of such applications include secure access to buildings, computer systems, laptops, cellular phones, and ATMs. In the absence of robust personal recognition schemes, these systems are vulnerable to the wiles of an impostor. In this paper we have developed and illustrated a recognition system for human faces using a novel Kohonen self-organizing map ( SOM ) or Self-Organizing Feature Map ( SOFM ) based retrieval system. SOM has good feature extracting property due to its topological ordering. The Facial Analytics results for the 400 images of AT&T database reflects that the face recognition rate using one of the neural network algorithm SOM is 85.5% for 40 persons.

http://www.airccse.org/journal/jcsit/1210ijcsit09.pdf

Pattern recognition using neural networks

In today's networked world, the need to maintain the security of information or physical property is becoming both increasingly important and increasingly difficult. Face recognition is one of the few biometric methods, which is very complicated system since the human faces change depending on their age, expressions etc. A human being has lots of expressions. So it is not possible to learn all types of expressions into the network. As a result, there is a solution for the unrecognized. Moreover due to the performance variations of the input device, face cannot be detected correctly and pattern may change extremely. In this paper we have developed and illustrated a recognition system for human faces using Kohonen self-organizing map (SOM). The main objective of our face recognition system was to obtain a model that is easy to learn i.e. minimization of learning time, react well with different facial expressions with noisy input and optimize the recognition as possible. Among the architectures and algorithms suggested for artificial neural network, the SOM has special property of effectively creating spatially organized “internal representation’ of various features of input signals and their abstractions. After supervised fine tuning of its weight vectors, the SOM has been particularly successful in various pattern recognition tasks involving very noisy signal. SOM are topologically ordered, which leads to good extracting feature ability. One develops realistic cortical structures when given approximations of visual environment as input, and is effective way to model the development of face recognition capability. The experimental result shows face recognition rate using SOM is 96.2% for 40 persons’ 400 images of AT&T database.

/pdf/neural-networks-for-face-recognition-using-som-2ledg781ul.pdf

Santaji Ghorpade

Papers

Pattern recognition using neural networks

Neural networks for face recognition using som