1,720,975 research outputs found
Set enumeration tree based image representation for gray level image storage and retrieval
No full textThe recent growth of communications and
multimedia applications had led to the requirement of mass
storage space as well as efficient retrieval technique especially
for multimedia data. In this paper, a novel approach for
representing gray level image for data storage and image
retrieval is proposed. The proposed approach used set
enumeration tree data structures where only unique image
pattern is stored in the image data structure. The overall
structure involves two types of tree data structuresthe first tree
is low-level image pattern tree to store the unique gray level
image pattern and the second tree is used to store the image path
by referring to the first tree data structure. The low-level image
pattern tree is predefined and will not expand throughout the
image encoding process. The size of the second tree is gradually
expanded as the result of addition of new image path during
image encoding. Through unique image pattern encoding into a
tree, there will be no redundant image features, thus leading to
saving storing space. Caltech-101 gray level image datasets were
used to test the proposed approach and the results showed that
it could lead to saving storage space while provide promising
performance in image retrieval
Optimization of image features using artificial bee colony algorithm and multi-layered perceptron neural network for texture classification
No full textOne of the fundamental issues in texture
classification is the suitable selection combination of input
parameters for the classifier. Most researchers used trial and
observation approach in selecting the suitable combination of
input parameters. Thus it leads to tedious and time consuming
experimentation. This paper presents an automated method for
the selection of a suitable combination of input parameters for
gray level texture image classification. The Artificial Bee Colony
(ABC) algorithm is used to automatically select a suitable
combination of angle and distance value setting in the Gray
Level Co-occurrence (GLCM) matrix feature extraction
method. With this setting, 13 Haralick texture features were fed
into Multi-layer Perceptron Neural Network classifier. To test
the performance of the proposed method, a University of
Maryland, College Park texture image database (UMD
Database) is employed. The texture classification results show
that the proposed method could provide an automated approach
for finding the best input parameters combination setting for
GLCM which leads to the best classification accuracy
performance of binary texture image classification
Skin Color detection Using Stepwise Neural Network and Color Mapping Co-occurrence Matrix
No full textSkin color has been proven to be a useful and robust cue for face detection, human tracking, image content filtering, pornographic filtering, etc. Most of skin classification researches are focused on using pixel-based method to classify skin and non-skin pixels. This paper proposed a new technique for region-based skin color detection using texture information. The texture information was extracted from the color mapping co-occurrence matrix (CMCM). This technique is extension of gray level co-occurrence matrix (GLCM) which is introduced by Haralicket. al to compute second order statistical texture features. The new color mapping matrix (CMM) between color bands have been developed for skin and non-skin area for each skin image and then, the CMCM were computed at four direction with distance, d = 1, and angle, θ = 0o, 45o, 90o, and 135o. The thirteen Haralick’s textures have been computed and used for formulating a skin color classifiers using stepwise neural network (SNN). The performance of each skin color classifier was measured based on true and false positive value. Besides that, the benchmark datasets from Universidad de Chile and TDSD were also be employed to test the skin color classifiers ability. The results shown that the skin color classifier formulated with [RGB] CMCM at direction (1, 0o) most superior as compared to other direction. Its average of true positive and false positive are 98.38 percent and 3.67 percent, respectively. Meanwhile, the classifier formulated with [RGB] CMCM at direction (1, 90o) is totally failed to classify skin and non-skin colors. Meaning that, the texture features which are computed from [RGB] CMCM at direction (1, 90o) cannot represent skin and non-skin color at all.</jats:p
The Use of Output Combiners in Enhancing the Performance of Large Data for ANNs
No full textDeriving classification information from large databases presents several challenges. The current methods used to classify a large dataset have the disadvantage of requiring long computational time and high complexity. In addition, most of the methods can only deal with selected features of the data while some of the methods can only deal with categorical or numerical attributes. This paper proposes large data solutions by defining the strategy to classify large data with local processors of Artificial Neural Networks (ANNs). A combination technique for reordered ANNs is proposed in modeling the combination of multiple ANNs as part of framework approach. Several repeated experiments with different techniques tested with the MNIST dataset show good percentage of performance and reduction of errors. The results obtained are in line with the importance of good performance achieved with the use of combiner for a large data solution
Going Beyond Counting First Authors in Author Co-citation Analysis
Get PDFThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Machine learning with multistage classifiers for identification of of ectoparasite infected mud crab genus Scylla
No full textRecently, the mud-crab farming can help the rural population economically. However, the existing parasite in the mud-crabs could interfere the long live of the mud-crabs. Unfortunately, the parasite has been identified to live in hundreds of mud-crabs, particularly it happened in Terengganu Coastal Water, Malaysia. This study investigates the initial identification of the parasite features based on their classes by using machine learning techniques. In this case, we employed five classifiers i.e logistic regression (LR), k-nearest neighbors (kNN), Gaussian Naive Bayes (GNB), support vector machine (SVM), and linear discriminant analysis (LDA). We compared these five classfiers to best performance of classification of the parasites. The classification process involving three stages. First, classify the parasites into two classes (normal and abnormal) regardless of their ventral types. Second, classified sexuality (female or male) and maturity (mature or immature). Finally, we compared the five classifiers to identify the species of the parasite. The experimental results showed that GNB and LDA are the most effective classifiers for carrying out the initial classification of the rhizocephalan parasite within the mud crab genus Scylla
Variations on the Author
Get PDF“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
- …
