One of the challenges in Content-Based Image Retrieval (CBIR) is to reduce the semantic gaps between low-level features and high-level semantic concepts. In CBIR, the images are represented in the feature space and the performance of CBIR depends on the type of selected feature representation.Late fusion also known as visual words integration is applied to enhance the performance of image retrieval. The recent advances in image retrieval diverted the focus of research towards the use of binary descriptors as they are reported computationally efficient. In this paper, we aim to investigate the late fusion of Fast Retina Keypoint (FREAK) and Scale Invariant Feature Transform (SIFT). The late fusion of binary and local descriptor is selected because among binary descriptors, FREAK has shown good results in classification-based problems while SIFT is robust to translation, scaling, rotation and small distortions. The late fusion of FREAK and SIFT integrates the performance of both feature descriptors for an effective image retrieval. Experimental results and comparisons show that the proposed late fusion enhances the performances of image retrieval.

One of the challenges in Content-Based Image Retrieval (CBIR) is to reduce the semantic gaps between low-level features and high-level semantic concepts. In CBIR, the images are represented in the feature space and the performance of CBIR depends on the type of selected feature representation.Late fusion also known as visual words integration is applied to enhance the performance of image retrieval. The recent advances in image retrieval diverted the focus of research towards the use of binary descriptors as they are reported computationally efficient. In this paper, we aim to investigate the late fusion of Fast Retina Keypoint (FREAK) and Scale Invariant Feature Transform (SIFT). The late fusion of binary and local descriptor is selected because among binary descriptors, FREAK has shown good results in classification-based problems while SIFT is robust to translation, scaling, rotation and small distortions. The late fusion of FREAK and SIFT integrates the performance of both feature descriptors for an effective image retrieval. Experimental results and comparisons show that the proposed late fusion enhances the performances of image retrieval.

This work presents the joint use of Radio Over Fiber and optical chaos to investigate the secure ROF link. Merging the two technologies, optical chaos for physical layer communication security and Radio over Fiber creates new design issues which have been identified and studied in detail in this paper for both analog Radio Frequency/Intermediate Frequency and digitized data. A semiconductor laser diode is driven into chaotic region using direct modulation scheme and RoF signal is added by chaos message masking scheme. The chaotically masked signal is transmitted over an optical communication link to investigate the propagation issues and synchronization of chaos at the receiver. The transmitted chaos is synchronized at the receiver to unmask the signal by using subtraction rule. To investigate the performance of chaotic communication system for Radio over Fiber transmission, the figure of merits like Bit error rate, Quality factor, Eye Opening Penalty and Root-mean-squared phase jitter are studied for digital data and Signal to Noise ratio and Total Harmonic Distortion are studied for analog waveform to address the effects of link length and data rate/message bandwidth. INDEX TERMS Chaos, chaos message masking, radio over fiber, secure optical communication, quality of service.

A critical requirement in optical chaos based secure radio over fiber (RoF) system design is the ability to control center frequency, spectral bandwidth, power level and signature of chaos to submerge message with sufficient horizontal and vertical margins both in time and frequency domains. Once frequency domain masking is completely achieved, time domain masking is met automatically, the former being more stringent. In a direct modulated semiconductor laser, the three control parameters are bias current (Ibias), modulation current (Imod) and modulation frequency (ωa). It is found that Imod increases bandwidth and amplitude dynamic range of chaotic pulses. Ibias increases the cavity power and hence average peak amplitude of laser chaotic pulses. The modulation frequency increases the speed of overall cavity dynamics and hence is used to increase the bandwidth of chaos but a corresponding increase in bias and modulation currents is required to support high repetition pulses. The results show relationship between three control parameters (bias current, modulation current and modulation frequency) in a direct modulated semiconductor laser and optical chaos bandwidth using regression.

No abstract