فصلنامه پردازش علائم و داده ها
سال چهاردهم شماره 2 (پیاپی 32، تابستان 1396)

The Joint-up and cursive form of Persian words and immense variety of its scripts and also different figures of Persian letters which depend on their sitting positions in the words have turned the Persian handwritings recognition to an intense challenge. The major obstacle of most often recognition ways is their inattention to sentence contexture, causes utilization of a word with correct appearance within an incorrect sentence all when input word is misrecognized. Sketching a solution that provides suitable analysis of sentence contexture requires huge linguistic resources to takes place as a fine representative for the chosen language to be recognized. In this article, a new method for Persian words online recognition is presented which tries to improve recognition process by using the term contexture. Also, to reduce the limits and rules that gainers compel to submit. The recognition method demonstrated in this article includes: the symptoms and morphemes framework of input handwritten are segregated and the framework of each morpheme with its symptoms is specified at first, then the symptoms of morphemes are specified and based on them a collection of words is being considered as a hypothesis. Each hypothesis is given a score by measuring its similarity to input handwritten and according to taken scores the likely hypothesizes are indicated. Then this procedure is led to achieve more likely hypothesizes by lingual model. To totalize the scores of a hypothesis, for the reason of the differences in scale of taken scores, a method of scores normalization is being offered. The test results demonstrate that by utilization of a language model with an online system of handwriting recognition, a significant reduction of words recognition error rate is being achieved. In addition of error rate reduction, by taking advantage of language model, a technique is being offered that can handle the Persian vocabulary recognition entirely. By availing the offered manner, the recognition precision at initial stage of letters level up to 95.9% and so the language model recognition up to 99.3% improved. So using a huge linguistic resources for Persian language and utilization of a language model, can improved the accuracy of recognition. For furture work, reinforcement learning algorithm is suggested for adaptation the algorithm to users.

Cardiac arrhythmias are one of the most common heart diseases that may cause the death of the patient. Therefore, it is extremely important to detect cardiac arrhythmias. 3 categories of arrhythmia, namely, PAC, PVC, and normal are considered in this paper based on classifier fusion using evidence theory. At first R peaks of ECG were identified. Then, the line features including ECG RMSSD, SDNN and HR Mean, and also non-linear characteristics were obtained by using SVD. The combination of these features results is given in MLP, Cascade Feed Forward and RBF neural networks. Next the principle of uncertainty about their response was checked, and finally, the results of these classifiers were combined by applying evidence theory. ECG processing is not needed to remove noise, however, the proposed method, in the presence of noise, is able to detect the cardiac arrhythmia, in best situation with 98% sensitivity.

Feature extraction performs an important role in improving hyperspectral image classification. Compared with parametric methods, nonparametric feature extraction methods have better performance when classes have nonnormal distribution. Besides, these methods can extract more features than parametric feature extraction methods do. Nonparametric feature extraction methods use nonparametric scatter matrices to compute transformation matrix. Nonparametric Discriminant Analysis (NDA) is one of the nonparametric feature extraction methods in which, in order to form nonparametric scatter matrices, local means of samples and weight function are used. Local mean is calculated by k nearest neighbors of each sample and weight function emphasizes on boundary samples in between class scatter matrix formation. In this paper, modified NDA (MNDA) is proposed to improve NDA. In MNDA, the number of neighboring samples, when measuring local mean, are determined considering position of each sample in feature space. MNDA uses new weight functions in scatter matrix formation. Suggested weight functions emphasizes on boundary samples in between class scatter matrix formation and focus on samples close to class mean in within class scatter matrix formation. Moreover, within class scatter matrix is regularized to avoid singularity. Experimental results on Indian Pines and Salinas images show that MNDA has better performance compared to other parametric and nonparametric feature extraction methods. For Indian Pines data set, the maximum average classification accuracy is 80.34%, which is obtained by 18 training samples, support vector machine (SVM) classifier and 10 extracted features achieved by MNDA method. For Salinas data set, the maximum average classification accuracy is 94.31%, which is obtained by 18 training samples, SVM classifier and 9 extracted features achieved by MNDA method. Experiments show that using suggested weight functions and regularized within class scatter matrix, the proposed method obtained better results in hyperspectrl imag classification with limited training samples.

In Natural Language Processing (NLP) studies, developing resources and tools makes a contribution to extension and Effectiveness of researches in each language. In recent years, Arabic Named Entity Recognition (ANER) has been considered by NLP researchers. While most of these researches are based on Modern Standard Arabic (MSA), in this paper, we focus on Classical Arabic (CA) literature. We propose a corpus called NoorCorp with 200k labeled words for research purposes which is annotated by expert human resources manually. We also collected about 18k proper names from old Hadith books as gazetteer which is called NoorGazet. Using ensemble learning, we develop a new approach for extraction of named entities (NEs) including person, location and organization. Adaboost.M2 algorithm, as implementation of multiclass Boosting method, is applied to train the prediction model. Results show that performance of the method is better than decision tree as the base classifier. We have used tokenizing, part of speech (POS) tagging, and base phrase chunking (BPC) to overcome linguistic obstacles in Arabic. An overall F-measure value of 86.85 is obtained. Finally, the proposed approach is applied on ANERCorp as MSA corpus and we have compared the results with NoorCorp.

Compression can be done by lossy or lossless methods. The lossy methods have been used widely than the lossless compression. Although, many methods for image compression have been proposed yet, the methods using intelligent skipping proper to the visual models has not been considered in the literature. Image inpainting refers to the application of sophisticated algorithms to replace lost or corrupted parts of the data so that visual difference cannot be inferred from the reconstructed image. In this paper, first we review some of the image inpainting algorithms and some of the image compression techniques using the inpainting algorithms, we propose a new inpainting based image compression algorithm that can improve the compression rate considerably. Simulation results show that our proposed method has reasonable visual quality in comparison with the other proposed image compression algorithms.

The degree of particles dispersion in the nanostructures is of the most important indicators used to verify the performance of the proposed methods in the synthesis of nanomaterials. Scanning electron microscopy (SEM) images of nanoparticles have structural, chemical and morphological information in nano-meters with high resolution. In this paper, classification of nanostructures using time series analysis and statistical features is presented. To reach this, SEM images were converted to the time series, and their statistical features were extracted from them. These features were fed to a fuzzy inference system as its inputs to classify the images of nanostructures as good, average and bad structures. The obtained results for 65 SEM images of nanoparticles with the same dimensions (250 × 250 pixels) showed more than 90% accuracy.

BCI P300 speller as a Brain computer interface system try to help disabled people and patients to regain some of their lost ability and help them to communicate via typing. The ability of personalization is one of the most important feature in BCI system. The typing language as a personalization factor is one of the most important feature in BCI systems. Most prior research on p300 spellers has focused on displaying English alphabet.in this research, we present a P300 Speller system, based on row or column paradigm, for Persian (Farsi) character input.as performance determination we calculated accuracy and bit rate for the said system based on recorded data from volunteers and reached the average accuracy of 91.39 % and bit rate of 7.22 (bits/minute) (we uses Linear LDA classifier for classification and the total trial number was set to 15). Furthermore in this research performance was measured for different trial number and final results demonstrated that this system can achieve high average accuracy of 82.82 % and average bit rate of 21.52 (bits/minute) by using only 4 repetitions.

Document images produced by scanner or digital camera, usually have geometric and photometric distortions. Existence of either type of distortion, deteriorate the performance of OCR systems. In this paper, we present a novel method to eliminate geometric distortion from document images. In the proposed method to eliminate the geometric distortion,first text lines are extracted from image, then each line is broken into equal-width columns. For each extracted segment from a line, its direction is corrected in such a way that the segment lies in horizontal direction. For this aim, for each different rotation of text segment, horizontal projection of its image is calculated and rotation which causes maximum of projection is considered as corrected direction of that segment and based on this, for each line segment parallel to horizon, a reference point, which is introduced as base direction, is extracted. Using reference points of each line segment, a polynomial is fitted to the text line. At the end, geometric distortion of each part of a text line is eliminated using a perspective transform which is estimated based on the extracted polynomial function. To increase the stability of the proposed method for short text lines, the curve fitting is performed using reference information for adjacent long lines. The proposed method is implemented on Persian and English databases and has been compared with the existing methods. The results indicate the efficiency and accuracy of the proposed method in elimination of geometric distortions.

Imperialist Competitive Algorithm (ICA) is considered as prime meta-heuristic algorithm to find the general optimal solution in optimization problems. This paper presents a use of ICA for automatic clustering of huge unlabeled data sets. By using proper structure for each of the chromosomes and the ICA, at run time, the suggested method (ACICA) finds the optimum number of clusters while optimal clustering of the data simultaneously .To increase the accuracy and speed of convergence, the structure of ICA changes. The proposed algorithm requires no background knowledge to classify the data. In addition, the proposed method is more accurate in comparison with other clustering methods based on evolutionary algorithms. DB and CS cluster validity measurements are used as the objective function. To demonstrate the superiority of the proposed method, the average of fitness function and the number of clusters determined by the proposed method is compared with three automatic clustering algorithms based on evolutionary algorithms.