Emir Sokic

Simple Computer Vision System for Chess Playing Robot Manipulator as a Project-based Learning Example

This paper presents an example of project-based learning (PBL) in an undergraduate course on Image processing. The design of a simple, low-cost computer vision system for implementation on a chess-playing capable robot is discussed. The system is based on a standard CCD camera and a personal computer. This project is a good tool for learning most of the course material that would otherwise be mastered by homework problems and study before an exam. An algorithm which detects chess moves is proposed. It compares two or more frames captured before, during and after a played chess move, and finds differences between them, which are used to define a played chess move. Further image processing is required to eliminate false readings, recognize direction of chess moves, end eliminate image distortion. Many Image processing problems and solutions can be introduced to students, through the proposed algorithm. The results are encouraging - students without any previous knowledge in image processing and advanced topics, such as artificial intelligence (neural networks etc.), may attain a chess move recognition success rate greater than 95%, in controlled light environments.

High performance disturbance observer based control of the nonlinear 2DOF helicopter system

This paper addresses the challenges of the disturbance observer (DOB) algorithms faced with highly nonlinear electromechanical systems which are dealing with high resolution and high speed operations. It describes the synthesis of robust and stable controllers and their applications in controlling azimuth and elevation angles of the helicopter model CE 150 supplied by Humosoft. Description of the helicopter, including its mechanical characteristics and mathematical model, is given in the paper. Tracking error, transient performances, power consumption and motor strains are used for the validation of control quality. Implementation of the control system on the experimental setup is also explained. MATLAB and Simulink are used as tools for developing the simulation model of the helicopter system. Obtained simulations are showing that developed controllers provide significantly improved results even in the presence of unknown and unpredictable inputs (disturbance and noise), unpredictable and unknown dynamics, external forces (torques) and change of the system parameters.

Novel fourier descriptor based on complex coordinates shape signature

Shape, color and texture are the most important discriminative elements for content based image retrieval. Fourier descriptors are widely used in shape based image retrieval problems. This paper presents a novel method of extracting Fourier descriptors from the simplest shape signature - complex coordinates. Instead of the commonly used scale normalization with the magnitude of the first harmonic, normalization with the sum of magnitudes of all harmonics is used. This leads to an improved shape scale normalization. All the experimental results indicate that the proposed method outperforms many other state-of-the-art Fourier descriptors based methods, both in terms of retrieval performance and computational time.

Analysis of off-line handwritten text samples of different gender using shape descriptors

The human experience in the analysis of the handwriting of male and female writers indicates that gender affects the appearance of the handwritten text. These differences are usually very difficult to describe numerically. In order to analyze the handwriting differences between male and female writers, several shape description techniques, such as the tangent angle function, curvature function and Fourier descriptors, were used in this paper. As an additional contribution of the paper, a database of 3766 off-line handwritten cursive and capitalized written words has been created. The database consists of male and female handwriting samples, classified by gender and handedness. The experimental results show that typical attributes of male and female handwritings imply certain differences in shape decriptors, and those differences have the potential for usage in gender handwriting discrimination.

Phase preserving Fourier descriptor for shape-based image retrieval

Shape is one of the most important discriminative elements for the content based image retrieval and the most challenging for quantification and description. Fourier descriptors are a very efficient shape description method used in shape-based image retrieval tasks. In order to achieve invariance under rotation and starting point change, most Fourier descriptor implementations disregard the phase of Fourier coefficients, consequently losing valuable information about the shape. This paper proposes a novel method of extracting Fourier descriptors that preserve the phase of Fourier coefficients. We introduce specific points, called pseudomirror points, and use them as a shape orientation reference. They facilitate the extraction of phase-preserving Fourier descriptors which are invariant under translation, scaling, rotation and starting point change. The proposed descriptor was tested on four popular benchmarking datasets: MPEG7 CE-1 Set B, Swedish leaf, ETH-80 and Kimia99 datasets. Performance and computational complexity measures indicate that the proposed method outperforms other state-of-the-art phase-based Fourier descriptors. In addition, it outperforms other state-of-the-art magnitude-based Fourier descriptors, and many non-Fourier based shape description methods in terms of performance - complexity ratio. HighlightsMagnitude-based Fourier descriptors discard information contained in phase.Invariant phase-preserving Fourier descriptor is proposed.Pseudomirror points are introduced, and used as shape orientation references.The proposed descriptor is compact, effective and simple to extract and compare.Retrieval performance is improved without increasing computational complexity.

Human-inspired shape-based image retrieval using weighted City-block distance and Fourier descriptors

Contour-based Fourier descriptors are very simple and effective shape description method used for content-based image retrieval. Similarity between Fourier descriptors is usually computed using measures such as City-block or Euclidean distance. These similarity measures consider all harmonics to be equally important, therefore harmonics with larger magnitude tend to have larger significance during the computation of shape similarity. In order to increase the importance of harmonics with lower magnitude, we propose to use weighted City-block distance for computing shape similarity. The proposed weighting coefficients are inspired by the contrast sensitivity of the human visual system to different spatial frequencies, known as the Contrast Sensitivity Function (CSF). Although weighted distances generally do not improve the retrieval performance, experimental results clearly demonstrate that human observers favour the retrieval system based on the weighted distances, and find it more accurate and relevant.

Shape description using phase-preserving Fourier descriptor

Contour-based Fourier descriptors are established as a simple and effective shape description method for content-based image retrieval. In order to achieve invariance under rotation and starting point change, most Fourier descriptor implementations disregard the phase of the Fourier coefficients. We introduce a novel method for extracting Fourier descriptors, which preserve the phase of Fourier coefficients and have the desired invariance. We propose specific points, called pseudomirror points, to be used as shape orientation reference. Experimental results indicate that the proposed method significantly outperforms other Fourier descriptor based techniques.

Synthesis of the integral sliding mode control and the robust internal-loop compensator for a class of nonlinear systems with matched uncertainties

This paper presents the design procedure of the integral sliding mode controller with enhanced robustness properties for a class of nonlinear uncertain systems. The integral sliding mode control (I-SMC) is synthesized with the generalized disturbance attenuation scheme called robust internal-loop compensator (RIC) through the Lyapunov redesign framework, thus introducing a generalisation of the well-known case for linear systems. The resulted two-layer control structure employs the classical controller with the feedforward term in the outer control loop to track the reference, while the inner control loop compensates the generalized disturbance and provides robust stability. The closed-loop system is proved to be asymptotically stable via Lyapunov stability theory. The developed control algorithm is used for attitude tracking of the small-scale helicopter system in the presence of additional parametric uncertainties and external disturbances. An excellent tracking performance and robustness stability of the proposed control method are revealed through computer simulations and experimental testing over the whole domain of the helicopter outputs.

Weighted-function based algorithm for retrieving handwriting trajectory from off-line data

Individuality of handwriting is the reason why it is used as a common base element for detecting character traits of the writer. It is believed that dynamic information improve the accuracy of the analysis, but they are not contained in an offline handwritten text. In order to recover dynamic information, a novel approach for handwriting trajectory recovery is proposed in this paper. The procedure is based on computing the objective function, which depends on parameters such as the angle of movement, path length, air pen tip movements, etc. The analysis is performed in MATLAB program package, using the text samples from IAM-OnDB database. The experimental results indicate that the average effectiveness of the proposed algorithm is above 75%.

A Hybrid Visualization of the Environment using Smart Ultrasound-based Sensor Arrays

This paper presents the work in progress on the design and testing of a distributed ultrasound-based sensory system for hybrid 1D and 2D environment visualisation. Many common sensors used in robotics, such as infrared and ultrasonic sensors, cameras and lasers mainly focus on quantifying distances and shapes, while rarely have the ability to differentiate among different sensed surfaces/materials. We propose an inexpensive prototype sensory system based on popular ultrasonic sensors which uses ultrasonic reflections to determine the acoustic reflection coefficients. This additional feature allows differentiating among sensed objects. Moreover, the developed ultrasonic cells are equipped with a microcontroller for basic signal processing and a communication link for integration into a sensor network. In this paper, we discuss the possibility of fusing obtained sensor array data and laser measurements.