Daniel Petrisor

Abstract-In this paper we present a new algorithm for pupil center localization. The face and eyes of the target subject are detected using the Cascade Classifier algorithm based on Haar-like features. A histogram equalization method is employed to increase image contrast. The core feature of our pupil center detection consists of an adaptive template matching procedure applied on the eye region image that integrates specific constraints related to location and colo r proprieties of areas surrounding the pupil in order to compute the precise pupil center. Experimental results demonstrate the efficiency of the algorithm by assessing the correctness of the pupil center detection with respect to a ground truth reference. This procedure will allow further development of a system that facilitates social interaction of impaired people by analyzing their eye movements and translating key zones of visual attention focus into sentences that alleviate their communication.

Abstract-The paper presents a cost effective pore water pressure sensor (PWP) for landslide prediction, based on an integrated silicon pressure sensor. The sensor can be associated with a landslide sensor or can work independently. The aim of the landslide sensor is to detect the wetting condition and the tiny displacements of the ground. The pore water pressure has a simple construction, by measuring the pressure of the air in a small enclosure, air that is compressed by the water table. The pressure is measured with a silicon pressure sensor and it is converted in mm water column. Depending on the underground structure, many such sensors are necessary at different levels. The information is collected at the ground level via a serial point to multipoint network. The sensor has a low cost, fine accuracy, low power consumption and is easy to install.

Abstract-The paper presents a device devoted to detect and measure earth displacements produced by landslides. It is part of inclinometer type geotechnical instruments and is based on measuring the deformation produced to a rod vertically mounted into the ground during soil layers sliding. With respect to other commercial devices, our inclinometer is characterized by high sensitivity to very small deformations owing to special strain gauges utilized in its construction, and also by possibility of 3D measuring, it being able to gauge in depth the amplitude and orientation of the soil layers displacement.

Abstract-The paper presents a device devoted to detect and measure earth displacements produced by landslides. It is part of inclinometer type geotechnical instruments and is based on measuring the deformation produced to a rod vertically mounted into the ground during soil layers sliding. With respect to other commercial devices, our inclinometer is characterized by high sensitivity to very small deformations owing to special strain gauges utilized in its construction, and also by possibility of 3D measuring, it being able to gauge in depth the amplitude and orientation of the soil layers displacement.

Abstract-The contrast enhancement is an important part of the image enhancement process. The visual details become more visible in an image with a high level of contrast. There are more techniques to improve a contrast image. In this paper we analyze the contrast modification by three main methods: the histogram equalization method, local histogram equalization (LHE) method and contrast limited adaptive histogram equalization (CLAHE) method. The degree of contrast enhancement is measured using the root-mean-square (RMS) parameter.

Abstract-The paper presents a device devoted to detect and measure earth displacements produced by landslides. It is part of inclinometer type geotechnical instruments and is based on measuring the deformation produced to a rod vertically mounted into the ground during soil layers sliding. With respect to other commercial devices, our inclinometer is characterized by high sensitivity to very small deformations owing to special strain gauges utilized in its construction, and also by possibility of 3D measuring, it being able to gauge in depth the amplitude and orientation of the soil layers displacement.

Abstract-The paper presents a study on the measurement stability of a landslide sensor based on four magnetic amorphous wire strain gauges used in an Wheatstone bridge configuration. The sensor is aimed for in-situ monitoring using a distributed sensor network. The experiment has been performed in the lab using a test stand, while a time step progressive deformation has been applied to the sensor. Results show a good stability of the measurement and a variable sensitivity with respect to the bending point.

Abstract-In this paper, we present a surveillance and measurement grid used for landslide risk assessment and mitigation. Landslide activity is measured using a landslide transducer node whose operation principle is based on strain gauges and controlled by an embedded system. Large susceptible risk areas are surveyed by our system using a grid of measurement points integrating multiple measurement nodes. Measurement nodes acquire displacement and orientation information reporting the data through RS485 to a microcontroller based unit coordinator. Measurement information from unit coordinators across the grid are transmitted wirelessly to a local server for storage and alerting actions. Analysis of landslide measured data shows that the land displacements measured with our transducer are correlated to the movements measured at the ground surface. Measurements regarding landslide displacement and orientation were used to implement a risk assessment and mitigation strategy concerning the hazardous outcomes of an imminent landslide.

Abstract-This paper presents a video processing framework developed in order to analyze eye and eyelid movements of impaired people. In the case of physically impaired individuals eye and eyelid movements may represent the only effective communication method that they could use to interact. Video processing algorithms are involved in the framework that we propose to aid people in this situation. A high speed camera is exploited to acquire the video sequences that will be processed by the framework in order to measure and analyze eye and eyelid dynamics.

Abstract-The paper presents the results obtained after performing an exhaustive study upon the properties of a special category of materials in form of magnetic amorphous microwires MAWs), in order to reveal their applicability to build sensitive elements for measuring strains and deformations. The study has been directed to the Stressimpedance (SI) effect, a variant of Giant Magnetoimpedance (GMI) effect that occurs in such materials. The goal of the study was to find the optimal conditions for developing a new type of strain gage in order to build a sensitive force measurement device that would be successfully applied in robotic arm control. Following the study, we built a strain gage possessing a gage factor of about 1000 times larger than that of a metallic gage, with a linearity under 1% for an operating range of ± 200 ppm.

Abstract-In this paper, we present the performance of a landslide transducer based on magnetic strain gauges and the interpretation of experimental results. Analysis of landslide data indicates that the land displacements measured with our transducer are correlated to the movements measured at the ground surface. Evaluating the landslide data, we have established a relationship between the system readings and the landslide movement. Measurements regarding landslide displacement and orientation were done using strain gauges based on amorphous magnetic microwires (MAW). In order to evaluate a larger area, our system uses a grid of measurement points containing multiple measurement nodes. Each measurement node acquires displacement and orientation information reporting the data through RS485 to the central unit coordinator. Measurement information from central unit coordinators across the grid is transmitted wirelessly to a server for storage and alerting actions.

Abstract-The paper presents a landslide transducer based on amorphous wire strain gauges. Four such strain gauges are mounted on a flexible tube and connected in a Wheatstone bridge powered with a 1 MHz sine wave. The voltages from the mid points of the two arms of the bridge are amplified and further peak detected. The transducer is build around a low power microcontroller. The voltages from the peak detector are converted into digital data and the displacement and the direction are computed and corrected. The result is sent via a serial interface toward a central unit. The originality of the paper consists of the use of such strain gauges as sensitive elements and processing of the signals for landslide detection. The results show higher sensitivity than classic strain gauges and the possibility to detect the direction of the displacement.

Abstract-This paper presents a video acquisition system used to record high speed video sequences in order to measure human eyelid mobility. The human eyelid mobility is a highly dynamic process which requires a high speed monitoring system. A high speed video camera is used to acquire video sequences of the subject face. The video sequences constituting the data base that are used to study eyelid mobility remain stored as binary files on the computer hard drive.

Abstract-A lane mark detection system is a driver assistance tool that automatically recognizes the lane mark painted on the road and the lane that he is travelling in. Such a system can warn the driver when has the tendency to move out from the lane, without being aware of it. This kind of system is a part of Advanced Driver Assistance Systems (ADAS), systems that has the goal to improve road safety. In this paper we propose an algorithm that automatically emphasizes the lane marks and recognizes them from digital images, by the use of Hough transform.

Abstract-The digital potentiometers are flexible devices allowing many of the applications of the mechanical trimming to be replaced by a solid state solution. In this paper, we address issues related to the use of the digital potentiometers in the active filter design, in order to implement programmable analog functions such as controlling the corner frequency and filter type. Also, we propose here a simple interfacing solution that allows for quickly configuring not only programmable filters, but other common circuits employing I2C digital potentiometers.

Abstract-This paper presents a virtual instrument for communication with a weather station that transmits information on measured environmental parameters using a radio telemetry units. The user interface is built in LabVIEW visual programming environment. This makes a PC serial or TCP/IP communication between a terminal and the central unit of the weather station, the information received is decoded and played back using the user interface program.

Abstract-The paper presents a high speed video processing algorithm designed to measure several parameters related to eyelid movements during blinking. First, the facial features of the subject are detected and tracked; then, the eye regions of interest are extracted using data regarding face orientation and eyebrows position. In the next stage, eye contour and representative lines for upper and lower eyelids are detected in each video sequence frame. Finally eyelid characteristics are extracted from the eyelid measurements data.

Abstract-A traffic sign recognition (TSR) system is a driver assistance tool that automatically recognizes traffic signs placed along the road. Such a system can disburden the driver by this task and can significantly increase traffic safety, by warning the driver about them. This system is part of Advanced Driver Assistance Systems (ADAS), systems that have the goal to improve road safety. In this paper, we present an experimental system that uses an Infrastructure to Vehicles (I2V) communication for recognizing the traffic signs. Using the Radio Frequency Identification (RFID) technology, it identifies the traffic signals on the road and, through a Human Machine Interface (HMI), the system warns the driver about it.

Abstract-This paper presents an image processing algorithm designed to measure eyelid movements during blinking. First, the facial features of the subject are detected and tracked; then, the eye regions of interest are extracted using information regarding face orientation and eyelids position. In the next stage, the eyelid line is detected from the image, after which the eyelid characteristics are extracted from the eyelid movement measurements.

Abstract-The paper presents an algorithm designed to face and facial features (eyes, eyebrows) detection. First, human skin regions are segmented and processed by morphological operators to detect the human face. After the face has been detected, a series of parameters that will be taken into account by the eyes detection stage are calculated. The most important parameters are face orientation and distance between the eyes.

Abstract-The paper proposes a three steps algorithm that automatically detects, classifies and recognizes traffic signs from images taken from a car running along European road. Traffic signs are detected by analyzing the color information contained in the images using HSV color space. Detected signs are then classified using correlation with standard sign shapes. The recognition step uses the minimum distance classification based on calculating the Euclidean distance between two feature vectors composed using a Gabor filter with different parameters.