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.
The key challenge of face recognition is to develop effective feature representations for reducing intra-personal variations while enlarging inter-personal differences. This paper presents a novel non-linear discriminant error criterion which can be used in effective feature learning from raw pixels. Unlike many existing methods which assume the problem to be linear in nature, the proposed method...
A functional neural network for handwritten digits recognition using single binary RRAM as the synaptic weight element, was presented, with successful recognition rate is up to 81%. The results show that multilevel or continual resistances is not necessary for resistive device used as synaptic weight element in hardware neural network application. Variation of RRAM devices, sometimes, would not affect,...
Recent advancement in unsupervised and transfer learning methods of deep learning networks has seen a complete paradigm shift in machine learning. Inspired by the recent evolution of deep learning (DL) networks that demonstrates a proven pathway of addressing challenging dilemmas in various problem domains, we propose a novel DL framework for expression-robust feature acquisition. The framework exploits...
In this paper we address the challenge of performing face recognition on human faces that are wearing glasses. This is a common problem for face recognition and automatic identity checking at airports, as passengers frequently forget to remove their glasses when passing through customs. In order to solve this problem, we first propose an automatic glasses presence detection model based on the tree-pictorial-structured...
In this paper, we implement a facial landmarking system to improve the performance of landmark location accuracy for the tree-structured based facial detector proposed recently by Zhu and Ramanan. Our main objective is to overcome their limitation where very small faces could not be detected and landmarked. Furthermore, we also want to improve the landmarking accuracy and reduce false positive rate...
Scalp electroencephalogram (EEG), a recording of the brain's electrical activity, has been used to diagnose and detect epileptic seizures for a long time. However, most researchers have implemented seizure detectors by manually hand-engineering features from observed EEG data, and used them in seizure detection, which might not scale well to new patterns of seizures. In this paper, we investigate...
In this paper, we aim to improve one of the current state-of-the-art models for facial components detection/localization. The objectives are to increase the amount of landmark points detected and improve the landmark extraction accuracy for frontal faces. The model is following Zhu and Ramanan's approach with a tree-structure. The popular AR dataset is chosen as an alternative training dataset as...
In this pilot study, we have (i) examined the relative importance of ten factors that can be used for developing new training methods and materials to improve employees' awareness and skills to defend against cyber risks, and (ii) investigated the relationship between an explicit security policy at the organizational level and individual employee's behavior and beliefs toward cybersecurity issues...
Grade classification of seed cotton is a major problem that has an significant impact on the agricultural economy. According to characteristics like impurities, yellowness and brightness that extract from images of seed cotton, constructing classification model of seed cotton base on the least square method. Using support vector machine regression to come up with a well improved algorithm. After full...
This paper presents an aerial image parsing approach. Firstly, scene categories of aerial image is extracted based on global feature-Spatial Envelope in the top-down step; secondly, the candidate proposals of object categories is classified in bottom-up step; finally, the bottom-up proposals are verified by top-down cues, and the parsing graph of the whole aerial image is obtained. The test on the...
In Chinese township enterprises, multi-dimensional piece-rate wages calculation are very common. Because of the diverse types of products and the multi-dimensional floating price, the automatic wages-calculation system is needed. Considering that the rapid development of markets, thus, it is very important that the wages settlement systems to be carried out the adjustment and optimization of dynamic...
The classical method of red cells examination in urine micrograph is counted by manpower, which has the following deficiencies such as low efficiency, strong subjectivity and poor reliability. This paper focuses on the detection of red blood cells in urine image captured under microscope by image processing. After the urine image is preprocessed by improved Sobel operator, red blood cells are localized...
Data of protein-protein interactions derived from High-throughput technologies are often incomplete and fairly noisy. Therefore, it is very important to develop computational methods for predicting protein-protein interactions. A sequence-based method is proposed by combining support vector machine and a new feature representation using Geary autocorrelation. SVM model trained with Geary autocorrelation...
In this paper, we proposed a two-stage hybrid model combined back-propagation (BP) neural networks and super-efficiency data envelopment analysis (DEA) model. It was applied to evaluate the performance of military maneuver engineering support. The indices of military maneuver engineering support were firstly built. And the CCR model was switched to super- efficiency issue. It will solve the disadvantage...
A new support vector machine (SVM) approach with fuzzy rules decisions is proposed for grout stratum identification, which integrates with SVM and fuzzy systempsila s merits. SVM is a novel machine tool and fuzzy Takagi-Sugeno model is easy to process uncertain system. Through defining lambda -sigmoid fuzzy function, we put forward SVM decision function rules-based expression and proved they are consistent...
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.