Ibrahim Furkan Ince

A new low-cost eye tracking and blink detection approach: extracting eye features with blob extraction

The systems let user track their eye gaze information have been technologically possible for several decades. However, they are still very expensive. They have limited use of eye tracking and blink detection infra-structure. The purpose of this paper is to evaluate cost effects in the sector and explain our new approach in detail which reduces high costs of current systems apparently. This paper introduces an algorithm for fast and sub-pixel precise detection of eye blobs for extracting eye features. The algorithm is based on differential geometry and still exists in OpenCpV library as a class. Hence, blobs of arbitrary size that means eye size can be extracted by just adjusting the scale parameter in the class function. In addition, center point and boundary of an eye blob, also are extracted. These describe the specific eye location in the face boundary to run several algorithms to find the eye-ball location with its central coordinates. Several examples on real simple web-cam images illustrate the performance of the proposed algorithm and yield an efficient result on the idea of low-cost eye tracking, blink detection and drowsiness detection system.

Execution Time Prediction for 3D Interactive CAPTCHA by Keystroke Level Model

CAPTCHA is a standard defense system which stands for Completely Automated Public Turing test to tell Computers and Human Apart. The basic idea of CAPCHA is providing questions that human users can easily recognize and answer but computer programs cannot. In this study, an interactive 3D CAPTCHA was introduced, and developed for better security. The design and evaluation phases are presented in detail. The proposed model passed the necessary security tests. The Keystroke Level Model (KLM) can be used to evaluate and predict task execution time for a specific system usage scenario. KLM is based on calculation on the sequences of the users’ tasks or actions. This study attempts to estimate execution times for 3D animated CAPTCHA by KLM. An interactive CAPTCHA method is developed by Flash Movie Technology which is used commonly in the web sites. The results show that performing the 3D CAPTCHA requires a serious amount of time. New interaction methods should be considered to improve the users’ performance.

Designing CAPTCHA Algorithm: Splitting and Rotating the Images against OCRs

Internet is being used for several activities by a great range of users. These activities include communication, e-commerce, education, and entertainment. Users are required to register regarding Website in order to enroll Web activities. However, registration can be done by automated hacking software. That software make false enrollments which occupy the resources of the Website by reducing the performance and efficiency of servers, even stop the entire Web service. It is crucial for the Websites to have a system which has the capability of differing human users and computer programs in reading images of text. Completely automated public Turing test to tell computers and human apart (CAPTCHA) is such a defense system against optical character recognition (OCR) software. OCR can be defined as a software which work for defeating CAPTCHA images and make countless number of registration on the Websites. This study focuses on a new method which is splitting CAPTCHA images into several parts with random rotation values, and drawing random lines on a grid background. Lines are in the same color with the CAPTCHA text and they provide a distortion of image with grid background. In this paper, the algorithm of our method is introduced in detail.

A user study: the effects of mobile phone prototypes and task complexities on usability

Mobile phones are increasingly becoming ubiquitous computational devices that are almost always available, individually adaptable, and nearly universally connectable (using both wide area and short range communication capabilities). Up to now, mobile phones have been used in purpose of the fact that we can send SMS messages among each other, save our important notes, save our schedules; connect to internet and so on. But how much can a mobile phone simplify our everyday interactions, before it itself becomes a usability burden? What are the capabilities and limitations of using mobile phones with different prototypes and complexities of the interaction items in itself. This paper presents a user study investigating the use of a prototypical, mobile phone based interaction system to send text messages and save schedules classified as simple and complex task settings. Our results show that mobile devices can greatly simplify our usage by using computer type prototypes with complex task settings. We have studied the effect of prototype of mobile phone and complexity of task for the usability test on subjective user evaluation. In this experiment, mobile phones with special software were given to test users (N=60) for a period of a week. Prototype design was done by using different kinds of platforms such as: paper, computer, and a fully operational appliance and also complexity of task was chosen as sending message and editing a saved schedule with the mobile phone, simple and complex task respectively. Later, a subjective user evaluation was measured in 5 points scale. Experiment results yielded that computer type prototype has significantly higher usability rate than paper type prototype and there is a significant interaction between prototype and complexity. Eventually, computer type prototype with complex task yielded highest rate while paper type prototype with simple task yielded lowest usability rates.

Patch-wise periodical correlation analysis of histograms for real-time video smoke detection

In this paper, an approach for video smoke detection is proposed. The basic idea is smoke has a highly varying chrominance/luminance texture in long periods. Since smoke has no shape, it also creates high shape changes in long periods. In this paper, two kinds of histogram are employed to observe change in luminance/chrominance texture and shape. Linearly interpolated chrominance/luminance subtraction image is used as input image for periodical analysis after thresholding. Intensity histogram which consists of 256 bins and oriented gradients histogram with 8 bins are employed for this purpose. Smoke generally creates transparent textures in which histogram bins create high variations. By considering the algorithmic cost and nature of smoke, periodical normalized cross-correlation analysis is performed in histogram bins instead of two-dimensional image context which makes algorithm more speedy and efficient for smoke detection. Experiments with a large number of smoke and non-smoke video sequences give promising results.

Classification of Geographical Features from Satellite Imagery

It is a challenging task to classify heterogeneous geographical features from satellite imagery. This paper addresses 31 straightforward classification algorithms based on predominantly pixels to classify miscellaneous geographical features from satellite imagery. The addressed algorithms can extract and process the features of a large dataset with high-resolution images expeditiously. A total of 606 red-green-blue satellite images of the Bosnian city of Banja Luka are exercised to comprehend their performances for classifying cemeteries, fields, houses, industries, rivers, and trees. The recorded experimental results demonstrate that the best average performance can come into possession of 87%.

Beyin Tümörü Tespitinde Görüntü Bölütleme Yöntemlerine Ait Başarımların Karşılaştırılması ve Analizi

Goruntu isleme teknikleri klinik karar destek sistemlerinde (KKDS) siklikla kullanilmaktadir. Bu calismada cagin onemli bir hastaligi olan beyin tumorlerinin goruntu isleme teknikleri ile tespit goruntulerinden (MRI) yararlanarak beyin tumorunun goruntu segmentasyonu ile tespit edilmesine yonelik bir calisma gerceklestirilmistir. Devlet hastanelerinden MR goruntuleri resmi izinlerle alinmis ve calismada kullanilmistir. Markov Random Field (MRF), Kapur, Kittler ve Otsu algoritmalari ile MR goruntulerindeki tumorlu bolgeler tespit edilmeye calisilmistir. Algoritmalar, MR goruntulerinin daha onceden belirlenmis bolgelerine (ROI – Region of Interest) ayri ayri uygulanmistir. Yapilan deneysel uygulamada Markov Random Field (MRF) algoritmasinin beyin tumoru tespitinde diger algoritmalara oranla daha basarili sonuclar verdigi gozlemlenmistir

CHILD AND ADULT CLASSIFICATION USING BIOMETRIC FEATURES BASED ON VIDEO ANALYTICS

As the number of social insecurity in regard to social crimes is on its rise, it requires a CCTV camera a higher accuracy in detecting the objects including pedestrians for efficient work of catching criminals. As the importance of the function of pedestrian detection is socially agreed upon, more studies on image and video based pedestrian detection have been conducted. In terms of that, the goal of this study is classification of pedestrian in two categories as a child and an adult. In this study, Haar cascade classifiers are used. This method first detects a full body and a head. Then, it measures the biometry given the relative proportioning length of a full body and a head. Moving average algorithm is used to obtain threshold ratio. Experimental results show the accuracy 100% for children and 64.5% for adults.

Uzun Mesafeli Hareketli Görüntülerde Yayaların Biyometrik Özelliklerine Göre Yaşlarının Sınıflandırılması

Video tabanli insan tespiti, gunumuzde bir hayli yaygin olan calismalardan biridir ve bu konu hakkinda bircok calismalar ve tasarilar mevcuttur. Daha da ayrintili bir calisma icin; nesne tespit edildikten sonra, nesneler siniflandirilabilir veya takip edilebilir. Yetiskin ve cocuk siniflandirmasi, sosyal guvenligin saglanmasi icin, ozellikle de gunumuzde artan vakalari goz onunde bulundurdugumuzda, pek yararli olabilir. Yapilan calismanin gayesi, videolardaki goruntulerden insanlari yetiskin ve cocuk olarak iki sinifa ayirmaktir. Oncelikle insan tespiti icin Haar siniflandirici kullanilmistir. Bir sonraki adimda ise, kafa ve tum vucut uzunlugu kullanilarak biyometrik bir oran cikarimi yapilmistir. Bu orana gore de tespit edilen insanin yetiskin veya cocuk oldugu belirlenmistir. Sonuclarimiz gostermektedir ki, yetiskin siniflandirmasindaki dogruluk payi %74.7 ve cocuk siniflandirmasindaki dogruluk payi ise %68.1’dir.

Investigation of New Statistical Features for BCI Based Sleep Stages Recognition through EEG Bio-signals

Electroencephalogram (EEG) is one of the oldest techniques available to read brain data. It is a methodology to measure and record the electrical activity of brain using sensitive sensors attached to the scalp. Brain’s electrical activity is visualized on computers in form of signals through BCI tools. It is also possible to convert these signals into digital commands to provide human-computer interaction (HCI) through adaptive user interfaces. In this study, a set of statistical features: mean entropy, skew-ness, kurtosis and mean power of wavelets are proposed to enhance human sleep stages recognition through EEG signal. Additionally, an adaptive user interface for vigilance level recognition is introduced. One-way ANOVA test is employed for feature selection. EEG signals are decomposed into frequency sub-bands using discrete wavelet transform and selected statistical features are employed in SVM for recognition of human sleep stages: stage 1, stage 3, stage REM and stage AWAKE. According to experimental results, proposed statistical features have a significant discrimination rate for true classification of sleep stages with linear SVM. The accuracy of linear SVM reaches to 93% in stage 1, 82% in stage 3, 73% in stage REM and 96% in stage AWAKE with proposed statistical features.