Sorin Zoican

Improved median filter for impulse noise removal

The paper presents a simpler and efficient median filter used for removing the impulsive noise from the images. A real time implementation using a digital signal processor (DSP) is also presented. The improved median filter presented has better performance than the classical median filter with no very great additional computational effort.

Intelligent System for Vehicle Navigation Assistance

Navigation assistance systems, aiming to improve safety and optimize the traffic, become more and more popular in vehicular technology. The reason is given by the significant traffic increase and congestion events in large cities, complexity of road infrastructure and unexpected or hazardous conditions that can be found on roads. This paper proposes an improvement in navigation systems by intelligently gathering traffic data provided by integrated car sensors and/or security systems and use these data to warn other participants. System concepts, architecture, preliminary design and performance evaluation aspects are presented, and also tasks organization for an embedded cost effective implementation using a real time kernel is illustrated.

Adaptive algorithm for impulse noise suppression from still images and its real time implementation

This paper presents a new algorithm to suppress impulse noise in still images, which allows removal of up to 90% noise. The paper is focused on two directions: performance evaluation of the new algorithm comparing with existing algorithms and evaluating the possibility to implement this algorithm in real time, for video clips, using the family of digital signal processors BF5xx signal (Analog Devices Inc.).

Audio signals noise removal real time system

This paper presents a real time implementation of a noise removal system for audio signals. This system is based on the wavelet approach with level dependant threshold that provides a useful method for noise elimination in the signals. Experimental results show that this method removes noise significantly and the system operates in real time.

Lane departure warning system implementation using the Blackfin microcomputer

the purpose of this paper is to evaluate the performance of an embedded lane departure monitoring system implemented using the Blackfin microcomputer. This paper is focused on evaluating the image processing algorithm and the main goal of this work is to propose methods to speed-up the necessary computations. A practical, flexible and low cost solution is proposed and the monitoring system is effective at medium vehicle speeds.

Computing packets priority in Diffserv networks using Cuda technology

The purpose of this paper is to evaluate the performance of GPU using in a router for a DiffServ-based network for video conferences using G729 and H264 standards for voice codec and video codec, respectively. The speech and video quality is improved by assigning a priority to each voice and video packet to the edge router of the network. The importance of a packet is determined differently for voice and video. A priority is assigned to the packet according to its importance, which ensures that important packets are more robust during transmission. The priority is computed using graphic processing units (GPU), in a parallel manner, using Compute Unified Device Architecture (CUDA) and therefore the performance is increased.

Speech de-noising system with non local means algorithm

This paper presents how the non-local means (NLM) algorithm may be involved in speech signals de-noising. This algorithm can be implemented in real time using a digital signal processor (DSP) such as ADSP2181 from Analog Devices. The performance of NLM algorithm is evaluated and compared with a wavelet based de-noising algorithm. Implementation issues are discussed in order to develop a real time system.

A nonlinear acoustic echo cancellation scheme implementation using the blackfin microcomputer

Nonlinearity of amplifiers produces nonlinear echo in acoustic systems, which degrades the performance of audio communication. The paper presents an acoustic cancellation scheme that involve the least mean square (LMS) algorithm to update the parameters of the nonlinear function and the variable step size NLMS (VSS-NLMS) algorithm to determine the coefficient vector of the FIR filter. The performance analysis and the real time implementation using a digital signal processor (DSP) microcomputer (Blackfin BF533, form Analog Devices) are illustrated.

Digital signal processing system for image pattern recognition

This paper evaluates the possibility of using a digital signal processor (DSP) in order to implement an image pattern recognition system based on a neural network architecture. The paper presents a brief introduction to neural network architectures and how such architectures can be used in pattern recognition; it presents the implementation of the neural network using a very powerful DSP microcomputer (eg, ADSP 2189 from Analog Devices), illustrates the main results for a character recognition system (execution time and error probability), and presents some conclusions.

Face Detection in Internet of Things Using Blackfin Microcomputers Family

This paper describes a face detection system based on the Blackfin microcomputer architecture that may be used in an Internet of Things (IoT) context. The face detection algorithm is based on skin detection and scanning binary images to determine the face area. Further image processing may determine the eyes and mouth in order to extract main face characteristics. The face detection algorithm may be used in context of IoT to determine the searching area for eyes and mouth (e.g. for face recognition and emotion detection). The face detection algorithm is implemented using the Visual DSP++ integrated development environment and face detection is achieved in real time.