There is a great deal of systems dealing with image processing that are being used and developed on a daily basis. Those systems need the deployment of some basic operations such as detecting the Regions of Interest and matching those regions, in addition to the description of their properties. Those operations play a significant role in decision making which is necessary for the next operations depending on the assigned task. In order to accomplish those tasks, various algorithms have been introduced throughout years. One of the most popular algorithms is the Scale Invariant Feature Transform (SIFT). The efficiency of this algorithm is its performance in the process of detection and property description, and that is due to the fact that it operates on a big number of key-points, the only drawback it has is that it is rather time consuming. In the suggested approach, the system deploys SIFT to perform its basic tasks of matching and description is focused on minimizing the number of key-points which is performed via applying Fast Approximate Nearest Neighbor algorithm, which will reduce the redundancy of matching leading to speeding up the process. The proposed application has been evaluated in terms of two criteria which are time and accuracy, and has accomplished a percentage of accuracy of up to 100%, in addition to speeding up the processes of matching and description.
Image is an important digital information that used in many internet of things (IoT) applications such as transport, healthcare, agriculture, military, vehicles and wildlife. etc. Also, any image has very important characteristic such as large size, strong correlation and huge redundancy, therefore, encrypting it by using single key Advanced Encryption Standard (AES) through IoT communication technologies makes it vulnerable to many threats, thus, the pixels that have the same values will be encrypted to another pixels that have same values when they use the same key. The contribution of this work is to increase the security of transferred image. This paper proposed multiple key AES algorithm (MECCAES) to improve the security of the tran
... Show MoreActive Magnetic Bearings (AMBs) are progressively being implemented in a wide variety of applications. Their exclusive appealing features make them suitable for solving traditional rotor-bearing problems using novel design approaches for rotating machinery. In this paper, a linearized uncertain model of AMBs is utilized to develop a nonlinear sliding mode controller based on Lyapunov function for the electromechanical system. The controller requires measurements of the rotor displacements and their derivatives. Since the control law is discontinuous, the proposed controller can achieve a finite time regulation but with the drawback of the chattering problem. To reduce the effect of this problem, the gain of the uni
... Show MoreIn the field of data security, the critical challenge of preserving sensitive information during its transmission through public channels takes centre stage. Steganography, a method employed to conceal data within various carrier objects such as text, can be proposed to address these security challenges. Text, owing to its extensive usage and constrained bandwidth, stands out as an optimal medium for this purpose. Despite the richness of the Arabic language in its linguistic features, only a small number of studies have explored Arabic text steganography. Arabic text, characterized by its distinctive script and linguistic features, has gained notable attention as a promising domain for steganographic ventures. Arabic text steganography harn
... Show MoreAlzheimer's disease (AD) increasingly affects the elderly and is a major killer of those 65 and over. Different deep-learning methods are used for automatic diagnosis, yet they have some limitations. Deep Learning is one of the modern methods that were used to detect and classify a medical image because of the ability of deep Learning to extract the features of images automatically. However, there are still limitations to using deep learning to accurately classify medical images because extracting the fine edges of medical images is sometimes considered difficult, and some distortion in the images. Therefore, this research aims to develop A Computer-Aided Brain Diagnosis (CABD) system that can tell if a brain scan exhibits indications of
... Show MoreA new de-blurring technique was proposed in order to reduced or remove the blur in the images. The proposed filter was designed from the Lagrange interpolation calculation with adjusted by fuzzy rules and supported by wavelet decomposing technique. The proposed Wavelet Lagrange Fuzzy filter gives good results for fully and partially blurring region in images.
In this research we will present the signature as a key to the biometric authentication technique. I shall use moment invariants as a tool to make a decision about any signature which is belonging to the certain person or not. Eighteen voluntaries give 108 signatures as a sample to test the proposed system, six samples belong to each person were taken. Moment invariants are used to build a feature vector stored in this system. Euclidean distance measure used to compute the distance between the specific signatures of persons saved in this system and with new sample acquired to same persons for making decision about the new signature. Each signature is acquired by scanner in jpg format with 300DPI. Matlab used to implement this system.
This research (The families of martyrs, victims of terrorism, war operations and military mistakes opinion about Al- Shuhada`a Establishment ) came to know and diagnose the mental image the martyrs victims of terrorism carries about the performance of Al- Shuhada`a Establishment and the services it provides to them, and to monitor the contents of that image that they had regarding their privileges and rights that Al- Shuhada`a Establishment supposed to give it to them, according to their law. The research problem was represented by the main question: (What is the image of Al- Shuhada`a Establishment among the families of martyrs, victims of terrorism, war operations and military mistakes), and this research was classified within the d
... Show MoreIn this study, we have created a new Arabic dataset annotated according to Ekman’s basic emotions (Anger, Disgust, Fear, Happiness, Sadness and Surprise). This dataset is composed from Facebook posts written in the Iraqi dialect. We evaluated the quality of this dataset using four external judges which resulted in an average inter-annotation agreement of 0.751. Then we explored six different supervised machine learning methods to test the new dataset. We used Weka standard classifiers ZeroR, J48, Naïve Bayes, Multinomial Naïve Bayes for Text, and SMO. We also used a further compression-based classifier called PPM not included in Weka. Our study reveals that the PPM classifier significantly outperforms other classifiers such as SVM and N
... Show MoreIts well known that understanding human facial expressions is a key component in understanding emotions and finds broad applications in the field of human-computer interaction (HCI), has been a long-standing issue. In this paper, we shed light on the utilisation of a deep convolutional neural network (DCNN) for facial emotion recognition from videos using the TensorFlow machine-learning library from Google. This work was applied to ten emotions from the Amsterdam Dynamic Facial Expression Set-Bath Intensity Variations (ADFES-BIV) dataset and tested using two datasets.
Recognizing speech emotions is an important subject in pattern recognition. This work is about studying the effect of extracting the minimum possible number of features on the speech emotion recognition (SER) system. In this paper, three experiments performed to reach the best way that gives good accuracy. The first one extracting only three features: zero crossing rate (ZCR), mean, and standard deviation (SD) from emotional speech samples, the second one extracting only the first 12 Mel frequency cepstral coefficient (MFCC) features, and the last experiment applying feature fusion between the mentioned features. In all experiments, the features are classified using five types of classification techniques, which are the Random Forest (RF),
... Show More