A.A. El-Harby

In this paper, two image compression systems are designed based on quadtree (QT). They can compress the colour images for the three components separately. The proposed systems divides colour images into their three components. Then, the first two components (R and G) are divided into blocks using QT method. While the division of the B component has the same blocks coordinates of the G component. The first system has three minimum values (MVs) and three difference values (DVs) for each block. In the second system for R component, one MV and one DV are identified for every block. While for the other two components, two MVs and one average difference (AD) are determined for any block. As a result, it is found that the division according to the G component is the best giving good compressed images with high compression ratios and visual quality. In addition to, the second system is the best one having the highest performance. This system has the highest accuracy rates in the compression...

The Square Scan Algorithm (SSA) was the first algorithm used to transform line features into vector representations automatically without any help from the user and without any use of other devices for instance, digitizer [1]. This paper presents some important modifications to the SSA algorithm. The main modification is done by drawing continuous squares to detect lines or branches without any lost segments. While the SSA draws each square after a distance called step length. The obtained performance was better and the processing time was reduced to less than the half approximately comparing to the previous SSA algorithm. The proposed algorithm may help with the integration of remotely sensed data and geographic information systems (GIS).

This paper aims to design three expert systems for recognising Unknown Chemical Radical (UCR). In addition, it determines how to measure an Intelligence Ratio (IR) for each system. The IR depends on Processing Time (PT) that is one of Natural Intelligence (NI) factors. The systems are designed to work in a systematic and efficient way. The third system was the best and the fastest one, as it can extend its Knowledge Base (KB) and have the ability to apply the required experiments without redundancy.

In this paper a new Natural Language Processing (NLP) system that has the ability to extract information from a database using rules and to search by the popular names. These names can be written as fixed forms or variant forms, for instance the name " راھیم 􀀾 اب " may be written correctly as " راھیم 􀀾 إب " or " راھیم 􀀾 اب " or wrong as " راھیم 􀀾 أب " for one person. The new search method is performed using Z-test, and we find that the new way has superior performance compared to the ordinary search.

The neural networks (NNs) simulation has been designed to simulate and predict the vortex wavelength ��x, lateral vortex spacing ��y, and normalized maximum vorticity at the vortex center near the wake of square cylinders with different corner radii. The system was trained on the available data of the three cases, although this data is very little. Therefore, we designed the system to work in automatic way for …finding the best network that has the ability to have the best test and prediction. The proposed system shows an excellent agreement with that of an experimental data in these cases. The technique has been also designed to simulate the other distributions not presented in the training set and predicted them with effective matching.

This paper presents a novel algorithm having two image processing systems that have the ability to compress the colour image. The proposed systems divides the colour image into RGB components, each component is selected to be divided. The division processes of the component into blocks are based on quad tree method. For each selection, the other two components are divided using the same blocks coordinates of the selected divided component. In the first system, every block has three minimum values and three difference values. While the other system, every block has three minimum values and one average difference. From experiments, it is found that the division according to the G component is the best giving good visual quality of the compressed images with appropriate compression ratios. It is also noticed, the performance of the second system is better than the first one. The obtained compression ratios of the second system are between 1.3379 and 5.0495 at threshold value 0.1, and b...

ABSTRACT This paper describes a new image processing algorithm for the automatic extraction of line features from raster images with width variation. The features are transferred into vector format. The algorithm has capabilities for extracting crossing and branching lines. In addition it has the ability to connect disconnected lines. The algorithm is implemented and also tested on several image cases

This paper presents an image compression algorithm that has the ability to divide the original grey level image into unoverlapped blocks depending on a threshold value. The proposed algorithm is based on quadtree. It can compress and decompress the image in easy way using two stacks instead of tree. In the compression process, the proposed technique stores the information of all blocks, for instance the upper left coordinate, size, minimum, and difference values in a stack, and the divided blocks are numbered in effective way. This information will be used to decompress the image again. It was found that the algorithm provides a high compression ratio ranged between 0.12 and 0.68.

The Square Scan Algorithm (SSA) was the first algorithm used to transform line features into vector representations automatically without any help from the user and without any use of other devices for instance, digitizer (1). This paper presents some important modifications to the SSA algorithm. The main modification is done by drawing continuous squares to detect lines or branches without any lost segments. While the SSA draws each square after a distance called step length. The obtained performance was better and the processing time was reduced to less than the half approximately comparing to the previous SSA algorithm. The proposed algorithm may help with the integration of remotely sensed data and geographic information systems (GIS).

ABSTRACT This paper presents two systems. They are an Adaptive Neuro-Fuzzy Inference System (ANFIS) and a neural networks (NNs) system that have the ability to simulate and predict the proton-proton (P-P) interaction. The P-P based ANFIS and NNs models calculate the multiplicity distribution of charged particles at different high energies. Simulation results of training charged particles using the ANFIS and NNs are tested with training data points showed perfect fitting to the data given by the experimental data. Prediction capability of the ANFIS and NNs are checked with data points not used in training proved also good performance. The results amply demonstrate the feasibility of these techniques in extracting the collision features and prove its effectiveness. It is found that, the ANFIS has got better performance and trained more quickly than the NNs system.

The neural networks (NNs) simulation has been designed to simulate and predict the vortex wavelength ��x, lateral vortex spacing ��y, and normalized maximum vorticity at the vortex center near the wake of square cylinders with different corner radii. The system was trained on the available data of the three cases, although this data is very little. Therefore, we designed the system to work in automatic way for …finding the best network that has the ability to have the best test and prediction. The proposed system shows an excellent agreement with that of an experimental data in these cases. The technique has been also designed to simulate the other distributions not presented in the training set and predicted them with effective matching.

This paper presents a novel algorithm having two image processing systems that have the ability to compress the colour image. The proposed systems divides the colour image into RGB components, each component is selected to be divided. The division processes of the component into blocks are based on quad tree method. For each selection, the other two components are divided using the same blocks coordinates of the selected divided component. In the first system, every block has three minimum values and three difference values. While the other system, every block has three minimum values and one average difference. From experiments, it is found that the division according to the G component is the best giving good visual quality of the compressed images with appropriate compression ratios. It is also noticed, the performance of the second system is better than the first one. The obtained compression ratios of the second system are between 1.3379 and 5.0495 at threshold value 0.1, and b...

This paper presents an image compression algorithm that has the ability to divide the original grey level image into unoverlapped blocks depending on a threshold value. The proposed algorithm is based on quadtree. It can compress and decompress the image in easy way using two stacks instead of tree. In the compression process, the proposed technique stores the information of all blocks, for instance the upper left coordinate, size, minimum, and difference values in a stack, and the divided blocks are numbered in effective way. This information will be used to decompress the image again. It was found that the algorithm provides a high compression ratio ranged between 0.12 and 0.68.

ABSTRACT This paper presents two systems. They are an Adaptive Neuro-Fuzzy Inference System (ANFIS) and a neural networks (NNs) system that have the ability to simulate and predict the proton-proton (P-P) interaction. The P-P based ANFIS and NNs models calculate the multiplicity distribution of charged particles at different high energies. Simulation results of training charged particles using the ANFIS and NNs are tested with training data points showed perfect fitting to the data given by the experimental data. Prediction capability of the ANFIS and NNs are checked with data points not used in training proved also good performance. The results amply demonstrate the feasibility of these techniques in extracting the collision features and prove its effectiveness. It is found that, the ANFIS has got better performance and trained more quickly than the NNs system.

... Al-Fatihah: The Opening 1. Bismi Allahi alrrahmani alrraheemi 2. Alhamdu lillahi rabbi alAAalameena 3. Alrrahmani alrraheemi 4. Maliki yawmi alddeeni 5. Iyyaka naAAbudu wa-iyyaka nastaAAeenu 6. Ihdina alssirata almustaqeema 7. Sirata allatheena anAAamta AAalayhim ...

An automatic system has been designed to classify fires and smokes from the Delta area in Egypt using Neural Networks (NNs). The proposed system is trained on the features of the provided image, and is designed to work in an automatic way for finding the best network that has the ability to have the best classification on data is not included in the training data. It is applied with varying number of layers and neurons in the hidden layers to determine the optimum network architecture. The system shows an excellent classification of the specified features, the obtained performance is 100% for test data that was collected from the training data, while is 99.6% for test data that was not included in the training data. The designed system can be carried out on remotely sensed images for classifying any other features.

This article presents two robust and efficient systems, they are based on the convolutional neural networks and the generative adversarial networks. Each one is used as a classifier to recognize the Libyan banknotes in effective way. The second system is used to generate artificial Libyan banknotes. The proposed systems were applied on three datasets using the same settings to have actual comparison. The first one was created by traditional methods involved 310 scanned images of real Libyan banknotes. While the second was generated by the second system based on the generative adversarial networks producing 540 images of Libyan banknotes. Each Dataset is divided into two divisions, one for training and the other for test. In the third dataset, some images from other banknotes were added to the test division of the second dataset. The two systems begin with extracting robust features from Libyan banknotes, and then comes the deep learning stage to create an intelligent classification....

An automatic system has been designed to classify fires and smokes from the Delta area in Egypt using Neural Networks (NNs). The proposed system is trained on the features of the provided image, and is designed to work in an automatic way for finding the best network that has the ability to have the best classification on data is not included in the training data. It is applied with varying number of layers and neurons in the hidden layers to determine the optimum network architecture. The system shows an excellent classification of the specified features, the obtained performance is 100% for
test data that was collected from the training data, while is 99.6% for test data that was not included in the training data. The designed system can be carried out on remotely sensed images for classifying any other features.

The Square Scan Algorithm (SSA) was the first algorithm used to transform line features into vector representations automatically without any help from the user and without any use of other devices for instance, digitizer [1]. This paper presents some important modifications to the SSA algorithm. The main modification is done by drawing continuous squares to detect lines or branches without any lost segments. While the SSA draws each square after a distance called step length. The obtained performance was better and the processing time was reduced to less than the half approximately comparing to the previous SSA algorithm. The proposed algorithm may help with the integration of remotely sensed data and geographic information systems (GIS).

This paper presents an automatic system that has the ability to restore diacritics (vowels) for non-diacritic Qur’an words, using a unigram base-line model and a bigram Hidden Markov Model (HMM). The proposed system was very robust and reliable without using morphological analysis methods for diacritics restoration. It was found that the HMMs are useful tools for the task of diacritics restoration in Arabic language. The used technique is simple to apply and does not require any language specific knowledge to be embedded in the model. Qur’an was used as corpora; our system was implemented and also tested on many parts of Qur’an as training set. For instance, the proposed system was implemented on 1366 words starting from the beginning of the Qur'an, and the best performance was 94.3% word accuracy for a unigram model and 95.2% word accuracy for a bigram HMM model.

This paper aims to design three expert systems for recognising Unknown Chemical Radical (UCR). In addition, it determines how to measure an Intelligence Ratio (IR) for each system. The IR depends on Processing Time (PT) that is one of Natural Intelligence (NI) factors. The systems are designed to work in a systematic and efficient way. The third system was the best and the fastest one, as it can extend its Knowledge Base (KB) and have the ability to apply the required experiments without redundancy.

This article presents two systems that have the ability to simulate and predict the proton-proton (P-P) interaction. They are an adaptive neurofuzzy inference system (ANFIS) and a neural networks (NNs) system. The P-P-based ANFIS and NNs models calculate the multiplicity distribution of charged particles at different high energies. Simulation results of training charged particles using the ANFIS and NNs as tested with training data points showed perfect fitting to the experimental data. Prediction capabilities of the ANFIS and NNs checked with data points not used in training also proved to perform well. The results amply demonstrate the feasibility of these techniques in extracting the collision features and prove their effectiveness. It is found that the ANFIS shows better performance and trained more quickly than the NNs system.

In this paper, two image compression systems are designed based on quadtree (QT). They can compress the colour images for the three components separately. The proposed systems divides colour images into their three components. Then, the first two components (R and G) are divided into blocks using QT method. While the division of the B component has the same blocks coordinates of the G component. The first system has three minimum values (MVs) and three difference values (DVs) for each block. In the second system for R component, one MV and one DV are identified for every block. While for the other two components, two MVs and one average difference (AD) are determined for any block. As a result, it is found that the division according to the G component is the best giving good compressed images with high compression ratios and visual quality. In addition to, the second system is the best one having the highest performance. This system has the highest accuracy rates in the compression ratios, peak-to-peak signal to noise ratio (PSNR) values, number of blocks and low computational time comparing with the first system.

This paper presents a novel algorithm having two image processing systems that have the ability to compress the colour image. The proposed systems divides the colour image into RGB components, each component is selected to be divided. The division processes of the component into blocks are based on quad tree method. For each selection, the other two components are divided using the same blocks coordinates of the selected divided component. In the first system, every block has three minimum values and three difference values. While the other system, every block has three minimum values and one average difference. From experiments, it is found that the division according to the G component is the best giving good visual quality of the compressed images with appropriate compression ratios. It is also noticed, the performance of the second system is better than the first one. The obtained compression ratios of the second system are between 1.3379 and 5.0495 at threshold value 0.1, and between 2.3476 and 8.9713 at threshold value 0.2.

Automatic recognition of vehicle plate characters became a very important in our daily life because of the unlimited increase of vehicles that enter or leave a supervised area. This paper describes an automatic intelligent system that captures the images of vehicles and has the ability to recognize the plates of vehicles entering the university campus or leaving. A camera is built on the university gate for taking images of incoming or outgoing vehicles. A rear image of a vehicle is captured and processed. The system applies an intelligent filtering of the input image based on a set of filters removing unnecessary image elements preserving the position and shape of characters of the vehicle plate, histogram manipulation and edge detection techniques for plate localization and characters segmentation. Adaptive Neuro Fuzzy Inference System (ANFIS) is chosen as a classifier for recognizing the characters in the vehicle plates. This system can extracts the letters and numbers from vehicles plates. So, the extracted vehicle characters can be used to record the incoming vehicles to the university campus and the outgoing. The recognized vehicle characters are saved in the university database.