Search

Image cloning which could copy the source image and “paste” it into the target image seamlessly is one of the important digital image manipulation techniques. Traditional methods which cope with this problem managed to analyze image information in gradient domains according to solving Poisson equations. As a result, existing methods need high computational cost and memory usage which hinder them from time-critical applications with high-resolution images. In this paper, we propose an algorithm to cope with image cloning problem based on modified mean-value coordinates technique so that we transform the problem into a very simple image interpolation procedure which is easy to implement without diminishing the performances. We also apply image pyramid technique to further accelerate the proposed algorithm and achieve high performances even on a CPU with single thread.

Image cloning which could copy the source image and “paste“ it into the target image seamlessly is one of an important digital image manipulation technique. Traditional methods to cope with this problem were to analyze image information in gradient domain according to solving Poisson equations. As a result, existing methods need high computational cost and memory usage which hinder them from real-time applications with high-resolution images. In this paper, we propose an algorithm to cope with image cloning problem based on modified mean-value coordinates technique so that we transform the problem into an image interpolation procedure which is very simple and easy to implement without diminishing the performances. Furthermore, we use GPGPU (General Purpose Computing on Graphics Processing Unit) technique to accelerate the proposed algorithm and achieve real-time performances with the cloned region whose size is about 1 mega pixels.

In this article, a high sensitive longitudinally driven giant magneto-impedance (LDGMI) effect in Fe 36 Co 36 Nb 4 Si 4.8 B 19.2 (FeCo-based) ribbon under weak magnetic field (WMF) was presented. The FeCo-based amorphous ribbon was prepared by single roller quench and annealed with joule-heat in flowing nitrogen atmosphere. The LDGMI effect profiles of FeCo-based ribbon have been measured with HP4294A impedance analyzer. The result shows that the LDGMI effect of FeCo-based ribbon responds to the WMF sensitively (as high as 1380 %/A·m−1). The super high sensitivity can be obtained in a moderate narrow range of annealing current density (27∼34 A/mm2) and closely depends on the driven current frequency. The highest sensitivity (1380 %/A·m−1) was observed when the FeCo-based amorphous ribbon annealed at 32 A/mm2 for 10 minutes and then driven with AC at the frequency of 300 kHz. The high sensitive LDGMI effect in WMF may results from the multiple magnetic anisotropy structure which induced by the temperature grads in the ribbon. This high sensitive LDGMI effect in WMF is promising for the development of weak magnetic-field sensors.

A colloidal dispersion of indium tin oxide (ITO) nanoparticles in butyl acetate solvent was achieved using a milling process in which PVP or PEG were used as a dispersant. The influences of Polyvinyl pyrrolidone (PVP) and Polyethylene glycol (PEG) dispersants on the stability of ITO suspension were studied by Zeta potential value, Nano-particle size measurement and sedimentation experiment. The results indicate that PEG dispersant dosage 5.5%, PVP 3.5% of the amount of slurry when the effect of dispersion can play, but in the storage time of 30 days, the dispersing agent PVP on the stability of slurry particles better than the PEG.

In this paper, a low-power CMOS variable gain amplifier (VGA) with an approximated exponential equation is presented. The proposed VGA is composed of two cascaded variable gain stages and a fixed gain stage, and has advantages of low-power, wide controlled gain range and temperature independent. The VGA is implemented in 0.18um CMOS technology. The simulations show that a total linear gain range of 80dB (−15dB∼65dB) with 3-dB bandwidth can be obtained at 50MHz; the IIP3 varies from −4dBm to −40dBm. The power dissipation is less than 2.2mA from a 1.8V supply voltage; the NF is only 14dB (at the 65dB gain). And the control circuit provides a gain error lower than ±1.5dB over the full temperature and gain ranges.

In this paper, a low-power CMOS variable gain amplifier (VGA) with dB-linear gain, which is based on an approximated exponential equation is presented. Being designed for GPS application, the proposed VGA is composed of two cascaded variable gain stages and a fixed gain stage, and has advantages of low-power, wide controlled-gain-range and highly-linear. The VGA is implemented in 0.18 ?m CMOS technology. The simulations show that a total linear gain range of 60 dB (27 dB-87 dB) with 3-dB bandwidth can be obtained at 16 MHz; the IIP3 varies from -11 dBm (at the minimum gain) to -60 dB (at the maximum gain). The power dissipation is less than 2 mA from a 1.8 V supply, and the input referred noise voltage is only 13 ?Vrms (at the 55 dB gain).

We present a multifunction wireless transceiver RFIC for WLAN and DVB-H applications. The RFIC includes a super-heterodyne transceiver for IEEE 802.11 a/b/g WLAN applications and a zero-IF receiver for DVB-H application. The WLAN and DVB-H transceiver share the down-conversion mixers, the baseband VGAs, filters, and the PLL synthesizer, resulting in a small silicon area of 20 mm2. Using a 3.3 V supply, the power consumption of the WLAN transceiver and the DVB-H receiver are 462 mW and 396 mW, respectively.

This paper presents a 5GHz double-balanced mixer with DC-offset cancellation circuit for direct-conversion receiver compliant with IEEE 802.11a wireless LAN standard. The analog feedback loop is used, to eliminate the DC-offset at the output of the double-balanced mixer. The test results show that the mixer with DC-offset cancellation circuit has voltage conversion gain of 9.5dB at 5.15GHz, noise figure of 13.5dB, IIP3 of 7.6 dBm, 1.73mV DC-offset voltage and 67mW power with 3.3-V power supply. The DC-offset cancellation circuit has less than 0.1mm2 additional area and 0.3mW added power dissipation. The direct conversion WLAN receiver has been implemented in a 0.35mum SiGe BiCMOS technology.

A low-cost low-power single chip WLAN 802.11a transceiver is designed for personal communication terminal and local multimedia data transmission. It has less than 130mA current dissipation, maximal 67dB gain and can be programmed to be 20dB minimal gain. The receiver system noise figure is 6.4dB in high-gain mode