In the evaluation and testing of the hardware-in-the-loop simulation and IR imaging systems, the dynamic IR scene projection is the key technology. With the successful development of the digital micromirror devices(DMD), close attention is paid to the dynamic infrared scene generation utilizing DMD. The application of dynamic IR scene projection to the simulation and testing of the IR imaging system are presented. The working principle of DMD is specified, and further discussion is also given to the core technologies and state-of-the-art of the DMD dynamic IR scene projection.

The design of infrared phase-shifting Twyman-Green interferometer is described. An automatic phase-shifting infrared interferometer whose aperture is Φ30 mm is developed. The computer software is programmed to fit the wavefront and evaluate the imaging quality of infrared optical system. The calculation accuracy of software equals to the similar softwares in the world. Several kinds of application of infrared interferometer--the quality of infrared optical system,the roughness of ground glass and the flatness of compact disk are reseached. The result show, the optical quality of the interferometric system is better than λ/25, the extended accuracy is better than λ/130, and the precision is better than λ/200(λ=10.6 μm).

An infrared wireless communication system is designed in 10Base-T Ethernet based on point-to-point infrared wireless communication. The wireless data communication in Ethernet is achieved through transforming Manchester code to 4PPM code and Ethernet data frames are re-packaged to be transmitted in physical layer by infrared wireless card. The transmission media of Ethernet is changed from unshielded twisted paired cable to infrared light. According to the character of infrared transceiver, optical collimation lens is applied in the infrared wireless communication system to extend the transmission distance of infrared transceiver.

The mixture of SF6 O2 and He which are regulated is injected into a discharge tube with 8 high-voltage electrodes. The divided F is injected into throat along flow direction. H2/D2 is injected along a long normal to the flow at the top and the bottom of the channel. The optical axis is located at 1 mm downstream of H2/D2 injectors. A small-scale continuous-wave DF/HF chemical laser is investigated. Experimental measurements indicate that its multiline output power is related to the flux of SF6, He and H2/D2. The effect on output power and laser device of O2 addition is investigated. The maximum net outputs of HF and DF are 12 W and 11 W respertively. The power is stable and regulated continuously, and the maximum operation time is 20 min.

As more and more laser threaten in the war field, the technology of laser warning is becoming an urgent task in the area of optoelectronic countermeasure in order to protect optoelectronic equipment. The basic technical requirement of laser warning and principle of radiation detection are discussed. Based on these theories, two ways to complete laser warning system are presented. According to our practical work, the detailed fulfillment of laser warning system using uncooled infrared focal plane array and experimental result are given.

The accuracy of measurement and detection probability of the sea surface using infrared laser in airborne laser bathymetry are discussed. The principle and analysis of the distance detection using infrared laser are introduced, and the accuracy of the inclined range at different scanning angles and different status of the sea, especially at different wind speed, is evaluated. Then the selection of discrimination threshold and the relationship between detection probability and SNR with giving criteria of detection probability are analyzed. Based on the simulation and parameters of the airborne laser bathymetry system, some improved methods of accurate measurement of inclined range are proposed.

An optical fiber magnetic-field sensor based on Faraday effect using BiCaVIG crystal is developed and the configuration of its sensing element is described. The experimental results indicate that its dynamic range is better than 70 dB with the nonlinearity less than ±5% and its noise equivalent magnetic flux density is 1.5μT. The sensor can be used to measure pulsed magnetic field. By the method of frequency spectrum analysis, the dynamic response characteristics of a kind of BiCaVIG crystal is measured and its -3dB bandwidth is about DC～200 kHz.

The method of measuring new first grade gauge block by phase shift interferometry has been studied. Firstly, multi-frame interference patterns are captured by CCD. Then the shape of the measured surface of the gauge and the assistant reference flat to be wrung are obtained by phase shift arithmetic. In particular, the unwrapping wavefront principle and technology when there are step and discontinuity in the gauge interference patterns are studied, and the discrete wave error is calculated to fit to a continuous and smooth surface. Measurement of guage block length is discussed, and measured result about dynamic argument of gauge block working face length is presented.

Wavelet transform is an important tool for image procession. The nonuniformity correction(NUC) is one of the key techniques in staring focal plane array imaging system. Based on the analysis of NUC of infrared focal plane array(IRFPA) by temporal highpass filter, the wavelet transform is used for the design of digital lowpass filter, which applies Mallat pyramidal algorithm to improve the frequency resolution of low frequency. The method is adaptive, which can efficiently overcome ghosting with proper scale. Its efficiency and surperiority are clarified by simulation.

The selection of matching area is an important element for the performance of the scene matching system.Through experiments,a three-dimensional model among SNR, repeated pattern and matching probability was erected.A new selection method based on a pre-selection via edge density and former model is presented. Finally the experimental result shows that this method can program the required regions successfully in certain conditions.

Several kinds of improved matching filter algorithms are proposed,which are optimal binary phase-only filter(O-BPOF) , amplitude encoded binary phase-only filter(AE-BPOF) and amplitude-compensated matched filter(ACMF). Three kinds of matching filter algorithms are compared and their respective superiorities are analyzed in different conditions.

OTDR is an important device used to automatically analyze the fiber's attenuation performance, but fiber's backscatter data has much noise, which has bad effect on fiber's events analysis. Wavelet transform has a good performance in both time and frequency field. A denoising analysis to the data using wavelet transform is given with better precision.

A kind of infrared image threshold denoising method is given. It combines wavelet transform with generalized cross validation. An asymptotically optimal threshold can be determined, without knowing the variance of noise, only using the known input data. After making discrete orthogonal wavelet transform to an infrared image, denoising is done in the high frequency subbands of each decomposition level respectively, so that the maximum signal-noise-ratio can be obtained in the high frequency subbands respectively. According to the experimental result ,the given algorithm can reduce the noise of infrared image effectively ,while it also keeps the detail information of infrared image well. As to performance and visual quality, the algorithm is better than the wavelet thresholding given by Donoho, the modified wavelet thresholding given by Johnstone and the traditional median value filtering method.

When the wavelet decomposition is used to denoise corrupted natural picture, the precision of variance estimation of wavelet coefficient has great influence on the denoising effect. Because the wavelet coefficients in different scales have quite different features, it is necessary to choose different window sizes in different scales to estimate the variance of coefficient in the window center. The outcome of variance estimation by maximal likelihood criterion is firstly analyzed. Then the variation and power of wavelet coefficient are also considered to determine the proper size of window in different scales. At last several simulations are performed to test the above analysis. The excellent simulated result shows the power of choosing proper window size in the specific scale.

The method for analyzing signal singularity using wavelet transform(WT), of which the prime function is smooth, is based on the relation of singularity and Lipschitz regularity of signals. The principle of the method is that the wavelet transform Wψ(a,t) of a signal equals to the first derivative of its smoothed version, s(t)θa(t), so that the maximum points of derivalive must occur at the sharply change points of the signal. Therefore, the maximum points of s(t)θa(t) must result in the singularity points of s(t). The simulation shows that the method can detect the singularity points of signals indeed.

A systematic and detailed comparison between two standards to characterize imaging IR system performance-minimum resolvable temperature difference/minimum resolvable contrast (MRTD/MRC) and triangle orientation discrimination threshold (TOD) is presented. Many differences between them, such as test pattern, measurement procedure, applicability and target acquisition(TA) performance prediction accuracy are described. Then the precise analysis is given for these differences. Finally, the application of TOD method in the performance evaluation and testing field of imaging IR system is prospected.

A merge-split algorithm based on study property of neural net and sum aggregate is presented. Its merit is that its merge is disconnected a little and freedom satisfies the adaptability, and it is realized during learning rather than stipulating in advance. Furthermore, the method of fitting edge of uncontinuous image with neural net is given. The method can carry on the fit at self-learning, using its learning function fully under the circumstances without the mathematical function form of edge. Some shortages of former marginal fit methods can be overcomed. 

Because there are many formats in video signals, it is hard to exchange information among them. It could be solved by the technologies, such as video format conversion. Based on discussing the algorithms of video format conversion, two algorithms with good effects, which are convenient for hardware implementation, are proposed. They are windowed function(with enhanced edge) algorithm and median filtering(with enhanced edge) algorithm. The first algorithm is simulated by software and tested on the FPGA board, and the results prove that the image has good quality and strong robustness. The second algorithm, by using the correlative information of space and time of the signal, also made the image get good quality with a sharper edge and lower hardware cost. Both algorithms have great practical value.

The development of real time hardware systems tends to be module-based as the systems become more and more complicated. Two methods of modules controlling and features are studied. Based on the research of general methods for modules controlling and the analysis of module functions of MPEG-2 encoder, a suitable method of controlling the modules of the encoder is presented and the hardware and software implementations of its controlling module are analyzed in detail. This controlling method is proved to be capable of guaranteeing the reliability of the real time system.

2D DCT plays an important role in image procession and video encoding. Based on all I-frame video encoding of MPEG-2, 2D DCT is implemented on FPGA. 8×8 2D DCT is changed into eight 1D DCT combination with some butter-fly operations and reordering operation. It is provided in simulation that the implementation of 2D DCT can run at 74.25 samples per second with only one 1D DCT module.

The method of image reconstruction uses some reasonable prior information to regularize the atmosphere turbulence degradation model (point spread function). The problem of calculation is transformed into optimization estimation of the criterion function with penalty term. The PSF values are estimated iteratively by ARTUR algorithm and then the images degraded by aero-optic effect are reconstructed. The quality of ARTUR algorithm is discussed and an experiment on image reconstruction is given.

Based on the principle of the lenticular screen covering color stereo-photograph and opticed theory of stereo imaging, a new composition method of the lenticular screen covering color stereo image based on single frame image is proposed. Instead of obtaining sequential images by expensive instruments, this method produces sequential images from only one image. Experiments indicate that the composed image is satisfactory and feasible and it has obvious advantage in economics and convenience.

By incorporating the local statistics of an image, a causal nonstationary autoregressive random field models can be extended to a nonstationary image. This nonstationary random field can provide a better description of the image texture than the stationary one. As a consequence of the model above as well as another model for labeling field, an image is better segmented. Due to low-order dependence in image for texture random field above, there are still false alarms. Therefore, high-order dependence as a new classification feature to improve the result of object detection is introduced. Entropy rate, depicting the feature, is estimated using random field model. The technique proposed above is applied to extract urban areas from landsat image.

In terms of the low SNR and contrast, the small and dim targets can not be easily detected in a single IR image, so utilizing a sequence of images is an effective method. Correlativity of target in a sequence of images is analyzed for detection of small and dim target. Some advanced algorithms are discussed, and concept of contrast correlativity is set forth. The method abstains from the disadvantage due to direct using gray correlation in the case that the gray is fluctuant in the sequence of images, and eliminates the disturbance of noise and background. The algorithm could be applied in the small target detection in other types of image.

The restoration of blurred image caused by rotation motion is studied. An algebraic restoration approach of deconvolving along blur path based on the least squares and space correlation constrain is proposed. Employing the optimal estimation can restore the pixels information in the corner effectively and rapidly. The algorithm is verified through the microcomputer and experimental results show that the proposed algorithm can remove space variant blur of rotation motion effectively.