Earticle

This paper investigates Farsi handwritten word recognition using common features. Also we applied biologically inspired features (BIFs), derived from a feed forward model of object recognition pathway in visual cortex for Farsi handwritten word recognition problem. Experimental results show that the model achieves high recognition percentage even for large variations and applicability of these features in Small Sample Size problems (SSS).The experiments were achieved using the Iranshahr dataset. This dataset consist of 780 samples of 30 city names of Iran which 600 samples for train and 180 samples for test was used. A set of experiments were conducted to compare Decision Templates with some combination rules. Results show that template based fusion method is superior to the other schemes.

This work proposes two main contributions to statistical steganalysis of Yet Another Steganographic Scheme (YASS) in JPEG images. Firstly, this work presents a reliable blind steganalysis technique to predict YASS which is one of recent and least statistically detectable embedding scheme using only five features, four Huffman length statistics (H) and the ratio of file size to resolution (FR Index). Secondly these features are shown to be unique, accurate and monotonic over a wide range of settings for YASS and several supervised classifiers with the accuracy of prediction superior to most blind steganalyzers in vogue. Overall, the proposed model having Huffman Length Statistics as its linchpin predicts YASS with an average accuracy of over 94 percent.

Sequence similarity in biological databases is used to characterize a newly discovered protein and confirming the existence of its homologs. This is often computationally very expensive. We have implemented a new algorithm that performs sequence similarity search using a pre-search phase. The proposed algorithm works in three phases. As a pre-preparation for Pre-Search, we locate a sequence, similar to the query sequence to extract all common words between the former and the latter. In the second phase, the pre-search phase, we locate all sequenes containing any of the randomly chosen common words. The list is further scanned in the third phase and the results obtained from the second phase are refined using Similarity Search (SS) algorithm, described in the paper. We have preprocessed the Uniref100.FASTA protein database containing 9,757,328 records downloaded from uniprot.org, to suit our application of sequence similarity search. The algorithm is simple and can be applied in various perspectives. These include searching in DNA and protein sequence databases, motif finding, and gene identification search. Pre-Search reduces the search space using much faster simpler algorithm. In large database search, its effect could be phenomenal.

In this paper, we made a comprehensive study to evaluate the performance of a multi-user MIMO MCCDMA wireless communication system. The channel encoded spatially multiplexed MIMO MCCDMA system under investigation incorporates four linear signal detection schemes (Equalizers) such as Minimum Mean Square Error (MMSE), Zero Forcing (ZF), Sphere Decoding and Q-less QR Decomposition under BPSK, DPSK, QPSK and QAM digital modulations. It is anticipated from the numerical results that with the Q-less QR Decomposition based signal detection scheme, the multi-user MIMO MCCDMA system outperforms in BPSK digital modulation under AWGN and Raleigh fading channels .In Zero Forcing detection scheme, the system shows comparatively worst performance. It has been observed from the present study that the system performance deteriorates with increase in order of digital modulation and noise power as compared to signal power.

This paper proposes a new classification method for Farsi handwritten word recognition using gradient and gradient based features. The extracted feature vectors were classified using two Multi Layer Perceptron networks as basic experts, and one Radial Basis Function was applied to choose the best expert. The experiments were performed using the Iranshahr dataset. This dataset consists of 780 samples of 30 city names of Iran out of which, 600 samples were used to train the network and 180 samples to test it. A set of experiments were conducted to compare proposed method with some other combination rules. Results show that the proposed method achieved 91.11% recognition rate.

This paper deals with a real world problem of medical diagnosis, to this goal, we propose to learn a compact fuzzy medical knowledge base through a cognitively-motivated granular hybrid neuro-fuzzy or fuzzy-neuro possibilistic model appropriately crafted as a means to automatically extract fuzzy weighted production rules. The main idea is to start learning from coarse fuzzy partitions of the involved proteins variations of input variables and proceed progressively toward fine-grained partitions until finding the appropriate partitions that fit the data. We provide details of implementation issues, experimental results, and discussion of interpretability issues. Moreover, learning is firmly grounded on fuzzy relational calculus, linguistic approximation and the crucial notion of importance widely used in human decision making and clinical problem-solving.