Convolutional neural network with joint stepwise character/word modeling based system for scene text recognition

Multimedia Tools and Applications - Text recognition in the wild is a challenging task in the field of computer vision and machine learning. Existing optical character recognition engines cannot...     

Rapid and efficient hand gestures recognizer based on classes discriminator wavelet networks

Multimedia Tools and Applications - The vision based on hand gesture recognition is one of the key challenges in behavior understanding and computer vision. It offers to machines the possibility of...     

Wavelet network for recognition system of Arabic word

Focusing on the development of new technologies of information, research in the speech communication field is an activity in full expansion. Several disciplines and skills interact in order to improve performance of Human Machine Communication Systems (HMC). In order to increase the performance of these systems, various techniques, including Hidden Markov Models (HMM) and Neural Network (NN), are implemented.     

A multiresolution wavelet networks architecture and its application to pattern recognition

This paper aims at addressing a challenging research in both fields of the wavelet neural network theory and the pattern recognition. A novel architecture of the wavelet network based on the multiresolution analysis (MRWN) and a novel learning algorithm founded on the Fast Wavelet Transform (FWTLA) are proposed. FWTLA has numerous positive sides compared to the already existing algorithms. By exploiting this algorithm to learn the MRWN, we suggest a pattern recognition system (FWNPR). We show firstly its classification efficiency on many known benchmarks and then in many applications in the field of the pattern recognition. Extensive empirical experiments are performed to compare the proposed methods with other approaches.     

Technological and chemical characters of bread prepared from irradiated wheat flour

The present work is a part of research programme aimed to use gamma irradiation to decrease the growth of fungi and the concentration of mycotoxines which present naturaly in collected samples of wheat flour from Egyptian markets (first part). To follow the technological, rheological and chemical characters of bread “Balady” common type in Egypt, same samples of wheat flour were irradiated with 2, 4 and 8 kGy for these purposes. The Egyptian bread type (Balady) had been prepared from irradiated and non-irradiated samples. The chemical analysis proved a significant induction in reducing sugars (RS) which increased gradually with increasing doses as 16.75%, 30.30%, 43.24% after using 2, 4 and 8 kGy respectively. Same trend was observed with total sugars (TS), less changes were observed in non-reducing sugars (NRS) and starch content. Results obtained showed that an actual reduction in dough development time, stability, weaking of dough and height of bread especially for irradiated samples at 4 and 8 kGy but 2 kGy was less effective. The reduction of Alkaline Water Retention Capacity (AWRC) was clear at high doses whereas 2.0 kGy improved the freshness of aged “balady” bread. Concerning, the evaluation of organoleptic characters of bread as judged by the panelists, all the samples were accepted either irradiated or not. But, the lowest values of acceptance resulted from irradiated samples with 4 and 8 kGy. Therefore, using 2 kGy can be recommended for keeping quality of wheat flour to avoid the infestation or pathogenic infection. Whereas high doses (4–8 kGy) can be use for eliminating the mycotoxins with some changes of quality which can be overcome by using some additional matters to get more acceptable bread and less changes in technological characters.     

A novel classification approach based on Extreme Learning Machine and Wavelet Neural Networks

Multimedia Tools and Applications - In this paper, we present a novel classification approach based on Extreme Learning Machine (ELM) and Wavelet Neural Networks. We introduce two novel...     

A deep stacked wavelet auto-encoders to supervised feature extraction to pattern classification

The major issue in pattern classification is in the extraction of features in the training phase. The focus of this work is on combining the ability of wavelet networks and the deep learning techniques to propose a new supervised feature extraction method to pattern classification. This new approach allows the classification of all classes of the dataset by the reconstruction of a Deep Stacked wavelet Auto-Encoder. This Network is obtained after a series of wavelet Auto-Encoders followed by a Softmax classifier at the last layer. Finally, a fine-tuning is applied for the improvement of our result using a back propagation algorithm. Our approach is tested with different image datasets which are the COIL-100, the APTI and the ImageNet datasets and is also tested with two other audio corpuses that contain Arabic words and French words. The experimental test demonstrates the efficiency of our network for image and audio classification compared to other methods.     

Development of a fuzzy logic traffic system for isolated signalized intersections in the State of Kuwait

In a conventional traffic signals controller, the lights change at constant cycle time. In many cities, automatic traffic signals are often based on a constant green-to-red cycle. The time period for green light (or red light) to be on is determined based on a stochastic model. The traditional vehicle-actuated control of isolated intersections attempts continuously to adjust green times. The decision to change green light duration involves fuzzy factors that cannot be precisely determined.The main objective of this paper is to develop a fuzzy logic traffic system that considers the two two-way intersections and is able to adjust changes in time intervals of a traffic signal based on traffic situation level.The proposed system has been applied and tested using real data collected from signalized intersection in Hawalli governorate in the State of Kuwait. Twenty-seven iterations have been done; the results show that the proposed fuzzy logic traffic system provides better performance in terms of total waiting time, total moving time, and vehicle queue. Finally, it can be observed from the results that the proposed system can be used to accelerate the cycle time and to give other phases the chance to gain more benefit from the green time lost.     

Natural occurrence of ochratoxin A in Tunisian cereals

Ochratoxin A (OTA) is a mycotoxin produced by several fungal species from Aspergillus and Penicillium genera. It is widespread in food and feed and its occurrence has been reported in cereals, cereal-derived products, dried fruits and spices. This mycotoxin was implicated in several human and animal pathologies such as the Balkan Endemic Nephropathy (BEN) and the Tunisian Chronic Interstitial Nephropathy (CIN) of unknown cause. In Tunisia, a clear correlation has been established between the consumption of OTA contaminated food and the induction of specific pathologies. Thereby, OTA was detected in human blood and tissues. The aim of our study was to investigate the presence of OTA in widely consumed cereals commercialized in Tunisia. The analytical methods used in our study involved the extraction of OTA by acidified toluene, immunoaffinity (IAC) clean-up and HPLC quantification with fluorescence detection. Levels and percentages of OTA contamination in different types of cereals, 110 wheat, 103 barley, 113 sorghum and 96 rice samples, were evaluated with incidences of 38%, 40%, 38% and 28%, respectively. The average of contamination by OTA found were 55, 96, 44 and 117 μg/kg, respectively, for wheat, barley, rice and sorghum. Our results showed that contamination percentages and levels in the period from 2004 to 2005 were higher then usual norms (5.0 μg OTA/kg) established by the European commission in 2002. The present report is the first one ever carried out on the natural occurrence of OTA in cereals, largely consumed by the Tunisian population.     

Evolution of ochratoxin A content during red and rose vinification

BACKGROUND: This work aims to investigate the effects of the common vinification steps on the fate of the ochratoxin A (OTA) during wine making. Two assays of red and rose microvinification, with artificially contaminated grapes, were performed. The content of this mycotoxin was also monitored throughout the process of red wine making from naturally contaminated grapes in a winery. RESULTS: The results from the different assays revealed that the maceration of pomace have a significant effect on the increase of OTA content in red wine (P < 0.05) whereas the alcoholic fermentation had a reducing effect. However, the spontaneous malolactic fermentation showed no significant effect on the OTA content in wine (P > 0.05). Storage of red wine in tanks followed by draining caused a significant decrease of OTA of about 55%. Clarification with a gelatin oenological fining agent contributed to the removal of up to 58% of OTA from red wine. CONCLUSION: Overall, a consistent decrease in OTA concentration was noticed throughout either red or rose vinification. This work has contributed to the understanding of the fate of OTA during different vinification processes, especially from naturally contaminated grapes. Copyright © 2008 Society of Chemical Industry