A fast and precise registration method for multi-image snapshot Fourier transform imaging spectroscopy is proposed. This method accomplishes registration of an image array using the positional relationship between homologous points in the subimages, which are obtained offline by preregistration. Through the preregistration process, the registration problem is converted to the problem of using a registration matrix to interpolate subimages. Therefore, the hardware interpolation of graphics processing unit (GPU) texture memory, which has speed advantages for its parallel computing, can be used to significantly enhance computational efficiency. Compared to a central processing unit, GPU performance showed ∼27 times acceleration in registration efficiency.
Fourier-transform imaging spectrometers are rapidly developed due to their extensive use in industrial monitoring, target detection, and chemical identification. Static Fourier-transform imaging spectrometer (SFIS) containing a birefringent interferometer is one of the most popular directions due to its inherent robustness. However, the SFIS suffers from its low achievable signal-to-noise ratio (SNR) because of the restriction of incident angle. Meanwhile, in applications, the SNR is perhaps the most important factor to determine the usefulness of an instrument. In this paper, we report here a Static Fourier-transform imaging spectrometer based on differential structure (SFIS-DS) in the 400-800nm wavelength range with a high SNR. As in electronic system, the differential structure can double optical efficiency and strongly restrain common mode error in the SFIS-DS. And the differential structure described here is also available for any instruments containing a birefringent interferometer. However, the drawback of the SFIS-DS is that the two images obtained by the two differential channels need precise registration which can be overcome by a sub-pixel spatial registration algorithm. The experimental results indicate the SFIS-DS can increase the SNR by no less than 40%.
This paper presents an equivalent direct detection receiver model by statistical method which simplifies the random
impulse responses of electrons counting of returned signal, background radiation and dark current as a Gaussian random
process with high enough gain. An investigation based on Gaussian distribution of system output in ICCD scannerless
range-gated Lidar system is conducted with the calculations of error probability, absolute error and relative error. As the
unique manipulated variable, optimized system gains are calculated separately based on the Gaussian model of the
random process to achieve the lowest error probability, the lowest absolute error and relative error. The simulations show
that the values of optimized gains tend to increase along with the target distance, although the increasing speeds are
different. To meet multiple requests, an evaluation model based on cost function is constructed with different weights.
The simulation shows that the evaluation model is capable of setting optimized gains for different circumstances and the
settings of the weights are vital to the performance of Lidar system.
Scannerless laser imaging radar will be the trend of laser imaging radar in future because it has several advantages of
high frame rate, wide field of view, small size and high reliability owing to giving up mechanical scanner. A scannerless
gain-modulated three-dimensional laser imaging radar is developed: Our system consists of a pulsed laser which is
capable of generating 100mJ pulses with a pulse width of 10ns and a center wavelength of 532 nm, and a receiver which
is a digital CCD sensor coupled to a GEN II intensifier with a 10nm bandwidth optical filter. The homogenized light
beam passes through a diverging lens to flood illuminate the targets. The return light is collected by a Nikon camera lens
and amplified by the image intensifier which is electronically driven and can be set to exponentially modulated gain or
constant gain. The CCD sensor can record a 12 bit gray-level image with a resolution of 780×582 pixels at a 50 Hz frame
rate. For a range image of the target can be extracted by processing an intensity image with exponentially modulated gain
and an intensity image with constant gain, the range image is acquired at a 25 Hz frame rate. During our outdoor
experiment, the range image of the targets at 500m is acquired with 2m range accuracy and the range image of the targets
at about 1 kilometer is acquired with 5m range accuracy in daytime.
The structure of DSC has a high vibration resistance and a high temperature resistance(up to 400..) and so, a DSC
sensor is adopted in the design of a vortex flowmeter, which is mainly comprised of a movable electrode and two fixed
electrodes. The movable electrode of DSC is a hollow cylinder and the fixed electrodes are two metal films
symmetrically plated on a dielectric cylinder. The air between the movable electrode and the fixed electrodes is the
medium of DSC. A Simplified Model is established to research the operation principle of Differential-Switched-
Capacitor. Based on the characteristics of vortices, a Differential Switched-Capacitor Converter is adopted, which utilize
a electronic analog switch to complete the charge and discharge of the two capacitors alternatively under the control of a
clock signal to convert the capacitance signal of DSC into a voltage signal. Experiments are performed on water medium
in a pipe of 25mm in nominal diameter. Five calibration points are selected and three tests are performed of each
calibration points. Experimental results show that the accuracy of DSC vortex flowmeter is 0.62% and the repeatability
error is 0.16%.
A fiber optic bending sensing system that uses twin-core fiber as the sensing element has been proposed and
demonstrated. The twin-core fiber act as a two-beam interferometer in which phase differences is a function of the
curvature, and it can be demodulated by the shift of the unique identification spectrum. By way of FFT analysis of the
white light interferometric spectrum, the variation of bending can be measured. The relationship between the bending
curvature and the shift of the unique identification spectrum has been given and the experimental results were also
A novel approach based on the theory of stochastic resonance (SR) for detecting weak signals in heavy noise for detecting distance information for laser ranging is presented in this paper. SR in a nonlinear system is a cooperative effect of noise and periodic signal driving in bi-stable systems. Under the proper condition, increasing input noise level results in an increase in the output signal-to-noise ratio (SNR), which means increasing the disorder of the input leads to increasing the order of the output. Driven by a periodic signal and a Gaussian white noise, stochastic resonance exists in the double-well potential system. This stochastic resonance phenomenon can greatly improve the SNR of a periodic signal with additive Gaussian white noise. In this paper the theoretical derivation for bi-stable system at the SR and the computer simulation have been given. Under the generic adiabatic approximation condition, a numerical simulation on such as the out SNR shows that the output SNR in heavy ground noise has been improved evidently. In laser ranging system, the SR theory was applied in electronic circuits and the out SNR improved obviously.
In optical remote sensing systems especially the imaging systems the quality of the scanner is the key element for obtaining high quality 2-D image of the target. There must be a detection device to verify the scanning performance of the optical scanner. In this paper such a detection device was developed. It consists of large area CCD camera, specific optical lens, electronics shutter, computer control element and scanning performance analysis software. The core technologies in developing the device are to design the specific optical lens, eliminate the delaying trail in the detecting image and recognize and analyze the spot matrix in the detecting image. The detection field of view can be wider than 10o×10o and resolution better than 0.1mrad. This device can not only perform real time detection on line, displaying dynamic values of all the detecting parameters, but also perform off line, displaying values of various detecting parameters for a given image taken at anytime. To consider the objectivity and creditability of the detecting result, there is an adjusting program to ensure the collimation between the detection device and the given optical scanner.
On-line real-time detection method for the defect of TFT-LCD is becoming increasingly important as TFT-LCD has replaced CRT displays and become the first choice in many applications. Traditional defect inspection methods of TFT-LCD are based on clear features and exact mathematic models. However, the defects of TFT-LCD are of strong complexity and vagueness. Moreover, determining the defects is a complicated process, which is influenced by the objective characteristics of the defects as well as the subjective factors of the observer. Therefore, it is very difficult to establish the accurate mathematical models for the defects. A fuzzy expert system approach is proposed for the defect inspection of TFT-LCD. Tests indicate that this system could emulate the experts or experienced operators to realize the automatization of the defect inspectin of TFT-LCD.