Koseoglu, Murat; Uyanik, Hakan

Audio search algorithms are used to detect the queried file in large databases, especially in multimedia applications. These algorithms are expected to perform the detection in a reliable and robust way within the shortest time. In this study, based on spectral peaks method, an audio fingerprint algorithm with a few minor modifications was developed to detect the matching audio file in target database. This method has two stages as the audio fingerprint extraction and matching. In the first stage, fingerprint features are extracted from spectral peaks on the spectrograms of audio files by hash functions. This state-of-art technique reduces the processing load and time considerably compared to traditional methods. In the second stage, fingerprint data of the queried file are compared with the data created in the first stage in the database. The algorithm was demonstrated, and the effect of spectrogram parameters (window size, overlap, number of FFT) was investigated by considering reliability and robustness under different noise sources. Also, it was aimed to contribute to new audio retrieval studies based on spectral peaks method. It was observed that the variation in the spectrogram parameters significantly affected the number of matchings, reliability and robustness. Under high noise conditions, the optimal spectrogram parameters were determined as 512 (window size), 50% (overlap), 512 (number of FFT). It was seen in general that the algorithm successfully detected the queried file in the database even in high noise conditions for these parameters. No significant effect of music genre was observed.

AtıfScopus9·WoS6·OpenAlex11·TR Dizin1

İncele

Gunay, Mihriban; Koseoglu, Murat

In this paper, at first, existing classification methods for classifying hand drawn circuits and circuit components are briefly introduced by examining advantages and disadvantages of each classification method. The handled methods are Support Vector Machine method (SVM), nearest neighbor method (KNN) and convolutional neural network method (CNN). Then, in the experimental part of the study, four different hand-drawn circuit components, which are diode, capacitor, resistor and ac voltage source, have been tried to be classified by using CNN method which is a highly recommended classification method in literature. The classification tables and graphics have been presented for the performed study. According to the obtained results and accuracy rate, it is concluded that CNN method can be used reliably to achieve high performance in the classification of hand-drawn circuit components.

AtıfScopus5·OpenAlex5

İncele

Uyanik, Hakan; Koseoglu, Murat

Window functions are structures that are frequently used to increase the quality of the transformation when processing the signal in the time-frequency domain, especially when creating the spectrogram of the signal. In calculations for frequency information, Fourier transforms of short-time samples of the signal are transformed into spectrogram data with the help of window functions, and then operations are performed by using these data. In this study, the performance of three different window functions, which are used in obtaining spectrogram data in the algorithm developed for audio fingerprint-based audio recognition, have been tested under different artificial noise conditions. The results of these three different window functions were compared graphically, and a truth table was obtained based on the analyzed database. As a result of the simulations carried out in MATLAB, it has been observed that the hanning window function gives better spectral results for different noise levels, and accordingly it produces better matching results compared to the other two window functions as reported in former studies on different applications. In addition, some improvements about window length have been added into algorithm to obtain better results.

AtıfScopus12·OpenAlex10

İncele

Gunay, Mihriban; Koseoglu, Murat; Yildirim, Ozal

In this paper, the Convolutional Neural Network (CNN) architecture, which is one of the deep learning architectures, is used to classify the basic circuit components drawn by hand. During the training and testing stages of the model, a new dataset containing images of 863 circuit components manually drawn by different people is created. The data set contains images of four different classes of circuit components such as resistor, inductor, capacitor and voltage source. All images have been fixed to the same size and converted to grayscale to increase recognition performance and reduce process complexity. In the study, training for four classes is performed with CNN architecture. Based on the CNN architecture, four new CNN models are employed with different the number of layers. The training and validation results of these models are compared separately, the model with the highest training and validation performance is observed with four layer CNN model (CNN-4). This model obtained 84.41% accuracy rate at classification task.

AtıfScopus16·WoS5·OpenAlex17

İncele