Building a system to identify individuals through their speech recording can find its application in diverse areas, such as telephone shopping, voice mail and security control. However, building such systems is a tricky task because of the vast range of differences in the human voice. Thus, selecting strong features becomes very crucial for the recognition system. Therefore, a speaker recognition system based on new spin-image descriptors (SISR) is proposed in this paper. In the proposed system, circular windows (spins) are extracted from the frequency domain of the spectrogram image of the sound, and then a run length matrix is built for each spin, to work as a base for feature extraction tasks. Five different descriptors are generated from the run length matrix within each spin and the final feature vector is then used to populate a deep belief network for classification purpose. The proposed SISR system is evaluated using the English language Speech Database for Speaker Recognition (ELSDSR) database. The experimental results were achieved with 96.46 accuracy; showing that the proposed SISR system outperforms those reported in the related current research work in terms of recognition accuracy.
Numeral recognition is considered an essential preliminary step for optical character recognition, document understanding, and others. Although several handwritten numeral recognition algorithms have been proposed so far, achieving adequate recognition accuracy and execution time remain challenging to date. In particular, recognition accuracy depends on the features extraction mechanism. As such, a fast and robust numeral recognition method is essential, which meets the desired accuracy by extracting the features efficiently while maintaining fast implementation time. Furthermore, to date most of the existing studies are focused on evaluating their methods based on clean environments, thus limiting understanding of their potential a
... Show MoreMost recognition system of human facial emotions are assessed solely on accuracy, even if other performance criteria are also thought to be important in the evaluation process such as sensitivity, precision, F-measure, and G-mean. Moreover, the most common problem that must be resolved in face emotion recognition systems is the feature extraction methods, which is comparable to traditional manual feature extraction methods. This traditional method is not able to extract features efficiently. In other words, there are redundant amount of features which are considered not significant, which affect the classification performance. In this work, a new system to recognize human facial emotions from images is proposed. The HOG (Histograms of Or
... Show MoreThe aim of this work, is to study color filming by using different intensities of fluorescent light, where we evaluate the capture image qualities for the RGB bands and component of L. And we study the relation between the means of RGBL values of the images as a function of the power of fluorescent light circuit . From the results, we show that the mean μ increases rapidly at low power values, then it will reach the stability at high power values.
Some degree of noise is always present in any electronic device that
transmits or receives a signal . For televisions, this signal i has been to s the
broadcast data transmitted over cable-or received at the antenna; for digital
cameras, the signal is the light which hits the camera sensor. At any case, noise
is unavoidable. In this paper, an electronic noise has been generate on
TV-satellite images by using variable resistors connected to the transmitting cable
. The contrast of edges has been determined. This method has been applied by
capturing images from TV-satellite images (Al-arabiya channel) channel with
different resistors. The results show that when increasing resistance always
produced higher noise f
Cyber security is a term utilized for describing a collection of technologies, procedures, and practices that try protecting an online environment of a user or an organization. For medical images among most important and delicate data kinds in computer systems, the medical reasons require that all patient data, including images, be encrypted before being transferred over computer networks by healthcare companies. This paper presents a new direction of the encryption method research by encrypting the image based on the domain of the feature extracted to generate a key for the encryption process. The encryption process is started by applying edges detection. After dividing the bits of the edge image into (3×3) windows, the diffusions
... Show MorePan sharpening (fusion image) is the procedure of merging suitable information from two or more images into a single image. The image fusion techniques allow the combination of different information sources to improve the quality of image and increase its utility for a particular application. In this research, six pan-sharpening method have been implemented between the panchromatic and multispectral images, these methods include Ehlers, color normalize, Gram-Schmidt, local mean and variance matching, Daubechies of rank two and Symlets of rank four wavelet transform. Two images captured by two different sensors such as landsat-8 and world view-2 have been adopted to achieve the fusion purpose. Different fidelity metric like MS
... Show MoreA system was used to detect injuries in plant leaves by combining machine learning and the principles of image processing. A small agricultural robot was implemented for fine spraying by identifying infected leaves using image processing technology with four different forward speeds (35, 46, 63 and 80 cm/s). The results revealed that increasing the speed of the agricultural robot led to a decrease in the mount of supplements spraying and a detection percentage of infected plants. They also revealed a decrease in the percentage of supplements spraying by 46.89, 52.94, 63.07 and 76% with different forward speeds compared to the traditional method.
Compressing an image and reconstructing it without degrading its original quality is one of the challenges that still exist now a day. A coding system that considers both quality and compression rate is implemented in this work. The implemented system applies a high synthetic entropy coding schema to store the compressed image at the smallest size as possible without affecting its original quality. This coding schema is applied with two transform-based techniques, one with Discrete Cosine Transform and the other with Discrete Wavelet Transform. The implemented system was tested with different standard color images and the obtained results with different evaluation metrics have been shown. A comparison was made with some previous rel
... Show More