Abstract-In this work, an automatic gender and age recognizer from speech is investigated. The relevant features to gender recognition are selected from the first four formant frequencies and twelve MFCCs and feed the SVM classifier. While the relevant features to age has been used with k-NN classifier for the age recognizer model, using MATLAB as a simulation tool. A special selection of robust features is used in this work to improve the results of the gender and age classifiers based on the frequency range that the feature represents. The gender and age classification algorithms are evaluated using 114 (clean and noisy) speech samples uttered in Kurdish language. The model of two classes (adult males and adult females) gender recognition, reached 96% recognition accuracy. While for three categories classification (adult males, adult females, and children), the model achieved 94% recognition accuracy. For the age recognition model, seven groups according to their ages are categorized. The model performance after selecting the relevant features to age achieved 75.3%. For further improvement a denoising technique is used with the noisy speech signals, followed by selecting the proper features that are affected by the denoising process and result in 81.44% recognition accuracy.Index Terms-Age classification from speech, gender classification from speech, MFCC based gender and age recognition, SVM classifier.

Received signals contain a vast amount of uncertainty due to the unknown modulating signals, communication channel, and noise. Therefore the modulation classification problem has to be approached based on artificial neural networks . In this work a digital modulation classification method is presented, based on discrete wavelet transform (DWT) and artificial neural networks (ANN) to distinguish digital modulation, like quadrature amplitude (QAM), phase shift keying (PSK), and frequency shift keying (FSK) signals. Feature extraction is performed via the DWT detail coefficients of the digital signals using (db4) mother wavelet, because of the usefulness of wavelet in signal de-noising . The extracted features are presented to an ANN for pattern recognition. In this work Levenberg- Marquardt error back propagation algorithm is used since it appears to be the fastest method for training moderate-sized feed forward neural networks (up to several hundred weights).The performance of the classification scheme is investigated through simulations using matlab-7, high recognition rates are obtained of about (97%). However, there are probabilities of misclassification of about (3%).

In this paper a text independent speaker recognizer from controlled noisy speech signals has been investigated. A recorded data is used for 20 Kurdish speakers (10 males, and 10 females) . The feature used in this work is the MFCC, and k-NN is used as a classifier. The recognition performance from the noisy speech signals has been improved by a denoising technique using wavelet transform. The result show that the de-noising technique could improve the performance of speaker recognizer by about 36%. KeywordsSpeaker recognition in noisy environment, MFCC features, k-NN classifier, de-noising signals in wavelet domain.

When using HD video cameras in stores a large storage space is needed. The required storage space can be decreased using motion and a color detection algorithm that makes the camera records only when a motion or a lighting change occur. So, in this work a real time motion and color detection is presented based on the difference of two successive frames of images after dividing the image into blocks, different block sizes are tested (2×2, 4×4 , and 8×8).YCbCr color system is used for each case with different thresholds for Cb, Cr, and Y. Good results for the motion detection can be obtained using Cb, and Cr other than Y, since the coloring components of an image are exist in Cr, and Cb, which are important for motion detection. Additionally the components that produce or reflect bright light can be obtained from Y. Consequently the result of lightening change detection at different illumination conditions is better using Y more than Cb, and Cr. This means that the best result for both motion, and lighting change detection is obtained by combining the chrominance (Cb, Cr) and luminance (Y) components of a video. The data used in this work are collected in an indoor environment using a laptop webcam under different illumination conditions.