Imam Samil Yetik

Distinguishing electronic devices using harmonic radar

A new approach to distinguishing electronic circuits using nonlinear/harmonic radar is presented in this paper. The radar transmits a single tone signal to the electronic circuits, consisting of nonlinear components, and exploits received harmonic response for separation. Unlike previous studies, the transmitted signal power is swept within a determined range so that the received powers at each harmonic are analyzed to capture the nonlinear characteristics. To predict the behavior of the nonlinear circuits such as diode clamper, diode limiter and full wave rectifier, certain statistical features of the received powers are analyzed. We show that the received power at the harmonics are distinctive for each circuit using Euclidean distances of features in feature space.

Hair region localization with optical imaging for guided laser hair removal

Laser hair removal is a popular nonsurgical aesthetic operation, where the aim is to remove unwanted hair permanently by damaging the hair follicle and shaft thermally. However, laser affects the superficial skin layers in addition to hair follicles causing health risks. Side effects of laser-assisted hair removal can be minimized by directing the laser beam only to the detected hair regions. This study proposes a feature-based hair region localization method using machine learning techniques, a first in this area. Features with low computational complexity have been proposed in order to discriminate hair and skin regions. Hair and skin region classification performances of different machine learning techniques have been applied and compared. Quantitative and visual results obtained from the proposed technique showed success in the detection of hair and skin regions. We concluded that the proposed method can be used in real-time guided laser hair removal devices.

Sparsity based off-grid blind sensor calibration

Abstract Compressive Sensing (CS) based techniques generally discretize the signal space and assume that the signal has a sparse support restricted on the discretized grid points. This restriction of representing the signal on a discretized grid results in the off-grid problem which causes performance degradation in the reconstruction of signals. Sensor calibration is another issue which can cause performance degradation if not properly addressed. Calibration aims to reduce the disruptive effects of the phase and the gain biases. In this paper, a CS based blind calibration technique is proposed for the reconstruction of multiple off-grid signals. The proposed technique is capable of estimating the off-grid signals and correcting the gain and the phase biases due to insufficient calibration simultaneously. It is applied to off-grid frequency estimation and direction finding applications using blind calibration. Extensive simulation analyses are performed for both applications. Results show that the proposed technique has superior reconstruction performance.

Radyoterapi uygulamaları için otomatik İris lokalizasyonu automated iris localization for radiotherapy applications

Uveal melanoma is a type of tumor that can cause loss of vision, loss of organ or even metastasis and loss of life. Radiotherapy is considered to be the least harmful and successful treatment type among various treatment methods. Radiotherapy should be carried out sensitively without movements of the iris. Therefore, the procedure is mostly performed by local anesthesia. Unfortunately, eye anesthesia can cause complications; therefore alternative methods are gaining importance. In this article, a method is proposed that can track the eye with a camera and automatically detect blinking so that radiotherapy can be aborted. Thus, we will be able to apply radiotherapy without anesthesia and it will be possible to stop the radiotherapy automatically so that the iris is not damaged in case of blinking. The developed method has been tested under various lighting conditions and it has been observed that the method has a very successful performance.

Distinguishing electronic devices using fourier features derived from harmonic radar

In this paper, a novel technique to distinguish between multiple types of electronic circuits using nonlinear radar is presented. The transmitter transmits a single tone signal to the targets, and harmonic responses are utilized to detect and distinguish non-linear electronic devices. Contrary to previous studies, the power of the transmitted signal is swept within a determined range. Then, first five harmonics of the received signal are analyzed. We utilize the information related to the frequency content using Fourier Transforms of the received harmonics are analyzed. We show that the energy levels of the Fourier Transforms can be used to classify various nonlinear electronic devices.

Morphological filter based graph cut for building detection

Automatic building detection with satellite and aerial images is a hot topic in remote sensing area. Especially, in recent years generation of high resolution Lidar data increased the success of building detection algorithms. In the proposed approach seed points in high elevation areas are obtained by applying morphological operations on Lidar data. Then, geodesic distances between this seed and other elevation points are calculated. Next, a mask is generated using distance map with shadow and vegetation information. Finally, this mask is given to the graph cut optimization and building detection is performed. Obtained results are compared with state-of-the-art algorithms. Results showed that our algorithm produced comparable precision and recall values with them.