The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
We propose in this paper a simple, yet efficient multi-channel fusion framework for joint acoustic event detection and classification. The joint problem on individual channels is posed as a regression problem to estimate event onset and offset positions. As an intermediate result, we also obtain the posterior probabilities which measure the confidence that event onsets and offsets are present at a...
This paper proposes a design method for an expert adaptive controller that stores operating data from experts in a database and designs a local proposal-integral-derivative (PID) controller at each step by referencing applicable data from this database. In recent years, the retirements of senior experts have created a serious social problem owing to the loss of their expertise. A skill-based PID controller...
In this paper, we propose models for emotion recognition from speech based on class-dependent features and Gaussian mixture models (GMM). Seven emotions are identified (Happiness, Fear, Neutral, Disgust, Anger, Boredom and Sadness) with a small set of features for each class. Results show that our system outperforms the single-stage classifier, with a 82.41% (74.86% in single-stage) overall recognition...
In NVR which is an NMR protein structure based assignment program, only N15 labeled data has been used until now due to the cost of acquiring C13 labeled data. Besides, in score function of NVR program, manually set parameters have been used for different proteins while calculating assignment probability of atoms. Nowadays, triple resonance experiments are standardized and obtaining C13 labeled data...
This study provides a new feature extraction method to gender classification. Local Zernike Moments is a method used for face recognition and proved that it is more successful than Gabor or LBP representations. In this study, LZM method is used for gender classification on FERET and LFW databases and demonstrated that it is more successful than LBP method on both databases. In the light of analysis...
In this study, it is aimed the determination of the masses and the masses in mammograms whether benign or malignant is intended to determine. Firstly; was carried out preprocessing on MIAS(Mammogram Image Analysis Society) database of international images. Mass separation was implemented with an appropriate threshold value. Then, Morphological filtering was performed for beter seggmentation. At classification...
In recent years, significant achievements have been achieved in the field of face recognition. Face recognition are special pattern recognition which are used in banking for identity approving and the entrance of controlled areas, the places where the security control impending to airports, to control machines, to follow-up of persons. In this study, face recognition applications on the Yale face...
In this study, the experimental studies were carried out on a database containing the types of wood knot. After preprocessing on the images in the database, specific features to knot were obtained using wavelet moments feature extraction algorithm. Type description is carried out with KNN classification algorithm by selecting most distinguishing the approximation coefficients on these features. In...
In this paper an eye localization algorithm which uses elliptical Gabor filters and local minima to detect eye pupils on a face image is proposed. First, face is detected in a given image. Then this face image is converted into gray-scale. Finally eye pupils are localized using elliptical Gabor filters and local minimum points. The method is tested on our combined data set using images form "ScFace",...
Feature extraction of music signals are often required in applications including; identification and classification of music files over the web or a database. The feature observations for different classes are expected to be distict, and also their computationally efficient implementations to be available. This is necessary both for offline and real time data processing. It is currently an active...
Joint picture and fusion of data that is acquired from different data sources is a complex and difficult process. One of the basic approaches in data fusion or data integration is schema matching. In contrast doing schema matching manually in many applications, it takes long time and many matching elements and situations are missed. It is very important to make schema matching automatically, since...
In this work human action recognition problem was discussed in video sequences. Solution of the problem was studied in three stages. Firstly, points of interest were detected with preproccesing and these points which are called cuboids were declared in small windows, then feature extraction was performed and finally, human action is decided by using classification. Features extraction is not only...
Local Zernike moments (LZM) is a recently proposed image representation scheme that is shown to be successful for face representation. In this study, a face verification system which depends on biometric hashing scheme, that uses LZM features extracted from face images is proposed. With the proposed system, security and user privacy is ensured. Verification performance of the system, in which user...
Traditional user authentication or identification systems are interested in something that you possess (like a key, an identification card, etc.) or something you already know (like a password, or a PIN). With biometrics, this interest has been shifted towards a different approach :something that are part of you (fingerprints or face) or something you make (e.g., handwritten signature or voice). Identification...
Colorization, the process of adding color to monochrome images, is a tedious and difficult task and often requires intensive manual effort by color experts. To alleviate this problem, a number of computational studies have been proposed in the literature which aim to perform this task in a relatively easy way, either by employing minimal user input in terms of color scribbles or using a colored reference...
Healthcare issues arose from population aging. Meanwhile, electrocardiogram (ECG) is a powerful measurement tool. The first step of ECG is to detect QRS complexes. A state-of-the-art QRS detection algorithm was modified and implemented. By the dedicated architecture design, the novel ASIC is proposed with 2.21 μW power consumption and 0.68mm2 core area. It is the smallest QRS detection ASIC so far...
Ambulatory physiological monitoring devices benefit patients, medical staff and hospitals by allowing patients to return home with the devices for monitoring. The main problem associated with designing such devices is that of power consumption. Wireless communications and complex processing are generally part of such devices and are power hungry components. These problems are magnified when dealing...
We propose a novel statistical approach for color texture modeling and classification based on cooccurrence matrices and discrete finite mixture models. Our statistical model assigns relevance weights to discrete cooccurrence features that are considered as random variables. Experimental results are presented to illustrate the merits of our approach on a difficult problem which is the categorization...
This paper proposed a two layer authorization mechanism, including traditional password system and rhythm recognition. The whole system includes two phases: preprocessing and usual operation for users. In preprocessing phase, users type password in a specific rhythm in order to record and analyse the characteristics of behaviour of users. In the second phase, how to verify a user in usual operation...
A new algorithm named wavelet transform weighted modular PCA is proposed for face recognition. Firstly, the training images and the testing image are preprocessed with wavelet transform and the LL band and the LH/HL average band are divided into sub-images with the same size. Secondly, the prospective classify contribution of each sub-model of the two bands are computed. Thirdly, each sub-image of...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.