Volume 1 Issue 1
Authors: Sorin Soviany; Cristina Soviany; Sorin Pu?coci; Mariana Jurian
Abstract: The paper proposes a classification method for people identification accuracy improvement in which the biometric system is trained not for all enrolled individuals but only for a few target identities to be recognized, therefore reducing the computational complexity for the large-scale biometric identification. The biometric detectors are relying on non-linear models which are more suitable for the real biometric data with high degree of intra-class variance; therefore they improve the people recognition accuracy even for the most difficult cases.
Keywords: Detector; Identification; Non-Linear
Authors: Lai. Y. P; Li. F; Zhou. Q. F; Huang. M
Abstract: Urban carbon emission has sharply increased owing to the rapid urbanization and motorization. Since transportation is a main source of carbon emission, it is urgent to build a low-carbon transport mode with characteristics of environment-friendly, high efficiency, low energy cost and safety. With the IPCC’s methods and spatial analysis, transportation carbon emission of Pingdi international low-carbon eco-park in Shenzhen was evaluated. Road vectorization was operated on arcgis10.0, while traffic distribution based on the dual-constrained gravity model was operated on transcad. Furthermore, by studying the structure of residents travel and the carbon emission per unit distance of each motor vehicle, the traffic carbon emission of the study area was evaluated. The total co2emission was 127127.57 tons per year. The transportation emission structure shows that private cars, official cars and taxi account for the largest proportion of the emission, each for 42.9%, 18.2% and 16.5%, while the proportions of bus and motorcycles were less than 10%. The spatial distribution shows that the largest intensity appears in roads of heavy traffic, such as Huiyan express and Yanlong Avenue. On the contrary, districts of more slip roads have low carbon emission intensity. influencing factors of transportation carbon emissions, including land use structure, structure of travel, transport vehicle, traffic network and low-carbon consciousness, were deeply analyzed in order to provide some references for developing the transportation mode dominated by public transport and the implementation of energy saving policy.
Keywords: Low-Carbon Eco-park; Transportation; Carbon Emission; Emission Structure; Spatial Analysis
Authors: Krzysztof Wójcik
Abstract: The main goal of this article is to describe briefly an image processing method utilizing the iterative OTO (Observation-Transformation-Operation) scheme, which creates the knowledge about images under consideration. The general knowledge structure, which may be regarded as a simple ontology, is built by the help of the inductive learning methods and some classes of knowledge transformations. The paper describes basic kinds of these transformations. The article presents an example of the practical usage of the proposed scheme in the chosen, relatively simple, task of image processing. In this context the proposed approach, which can be called “conceptual filtering”, may be considered as a variant of Syntactic Image Understanding methods. Additionally, the prospect of the application of the OTO scheme in various kinds of identification and classification tasks is proposed. As a conclusion, the article points to a main condition of a successful usage of the described methodology, namely reducing the size of the hypothesis space which may be done by the removal of the unused concepts and objects, and the usage of the operations of knowledge structure joining.
Keywords: Image Processing; Pattern Recognition; Machine Learning
Authors: Chunni Dai; Qingli Li; Jingao Liu
Abstract: Hyperspectral imaging is a relatively new method for identifying blood cells. Except for morphological and texture information in gray images, hyperspectral data contains a lot of spectral signatures which represent chemical analysis of a sample. Therefore, hyperspectrum has an advantage over digital color images due to spectral signatures. With these spectral and spatial features blood cells can be recognized and classified. Over 40 features are extracted from hyperspectral image sequence. These features include spectral pattern traits and similarity measures. To implement blood cell discrimination, a back propagation neural network (BPNN) is proposed in this paper. The connection weights of the BPNN were fixed through the training by a genetic algorithm (GA) which employed two adaptive mechanisms during the evolutional processes. Three-fold cross validation was applied to classify blood cells of given samples. Experimental results demonstrated that the classifier using a BPNN and an adaptive GA was effective. Finally, this paper also described a cursory investigation of the effect of spectral data volume on classification accuracy. Two compressed image series which can be viewed as multispectral series were obtained by systematic sampling from two original hyperspectral series, respectively. Compared to multispectral data, the hyperspectral data with high dimensionality achieved superior accuracy in recognizing blood cells, although requiring greater processing time due to the large amount of data dimension.
Keywords: Classification; Blood Cell; Hyperspectral Imaging; BP Neural Network; Genetic Algorithm