The sky background brightness has been measured with sky background brightness measurement equipment in desert region of northwest china, the spectrum distributing of background brightness of the region is researched, and the detecting ability of the Electro-Optical system is also researched and simulated through the filtering technology in the region under the spectrum distributing. The research results indicate that the background spectrum distributing is obvious different to the theory spectrum, the proportion of the 600～900 nm band is bigger relative to theory, and it varies slowly with the variety of azimuth angle and zenith angle. The detecting ability can only improve 0.1 magnitude through filtering technology when the background brightness is 10 W·m-2Sr-1.

Placing a mirror in front of the high-speed camera, turning it according to the discipline of moving projectile and relationship of imaging, the flying procedure of the projectile can be acquired. To ensure the real-time quality of this system, the control data can be calculated before real test start, and saved into memory cell. During the tracking in real time, the tracking control unit is used to dominate the mirror to complete the follower-photograph. Finally, this system can follow the whole process, the image of projectile is suitable for reaching the purpose of magnify the follower scope.

The composition and the working principle of the test system for the sapphire optical fiber temperature sensor were analyzed. As there is a big temperature difference between the inner and outside surface of black body cavity, so the error of temperature of the measured object is presented. To solve this problem, a process of separating variables was used to derive the outer surface temperature of black body cavity from the inner surface temperature of the blackbody cavity, which further improved the performance of the sensor required by test. The experiment tested the feasibility of this method, and the measurement accuracy of the sapphire optical fiber blackbody cavity temperature sensor was improved.

The GeoEye satellite system, the next generation electro-optic system, the STSS and the Tacsat system in American Space-based Electro-optic Reconnaissance System are studied. The mainly technology indexes and operation characteristics of high resolution, full color, multi-spectrum cameras, advanced wide FOV acquisition sensors, staring multi-band tracking sensors for narrow FOV are analyzed. The large diffraction film optic system, the advanced optic camera system, optic sensors, infrared detectors, visible light imagers and infrared imagers are briefly introduced in future development plans.

At present, anti-sniper EO detection includes two methods mainly, IR detection and laser detection. Taking U.S.A., France, Israel and Russia for examples, the developments of anti-sniper IR and laser detection technology and equipment are introduced. From the advancement of EO detection materials, integration of detections and countermeasures, expansion of installation platform，fusion of multifunction, realization of network capability, the trends of anti-sniper EO detection technology and equipment are discussed lastly.

When the flight velocity of a target is more than 5 Ma, its radiation peak wavelength falls within the range of shortwave infrared (SWIR), and with the distinctive characteristics such as stronger in radiated power and varying fast with time, it is easy to be detected by the SWIR sensor. SWIR detector transduces the incident photons of the target reflected by atmospheric particles into electrical signals. Through monitoring the amplitude change of electrical signals, a strong radiation source approaching can be estimated, and then the early warning time can be advanced.

A new method of measuring the illumination distribution of the laser spot was studied. The method adopts the LabVIEW to drive the USB camera to collect the image of the laser spot, gets the 3D gray distribution and the illumination distribution of the laser spot after having processed the image with the computer, and can observe timely the exact illumination value of every pixel. The parameters of system design software could be set flexibly. With this method the data information of the laser spot could be collected dynamically and be processed real-timely. This method had been used for experiments teaching.

In order to improve the recognizing ability of the infrared objectives under the island's background, the recognition method is proposed based on average gray. While the infrared detector recognizes a ship as a target at the long distance, it's too difficult to identify it from the surrounding near the ship and the islands which have the exterior contour similar to the ship. Through researching the islands and the ships' component, comparing their infrared characteristics and as well as getting the data from the IR image, it is found that the average gray of the ship is higher than the island's average gray.

The measure of warship wake bubble is a base for developing an underwater sound and light wake detector. The distribution characteristics of the warship wake were studied, and the difficulties in the bubble image measured by the image method were analyzed. A new method of bubble measure is put forward by use of high speed photography based line lighting lamp-house. The system of bubble controlling is established through the air pump and chinaware. The bubble image of different size is gained, and characteristic of bubble is analyzed based numeric image disposal. The result shows the measure of bubble characteristic by use of high speed photography based line lighting lamp-house is a good way.

With the increasing demand of communication, the microwave communication is attracting more and more attention. The MPF(Microwave Photonic Filter) is one of the most important parts in the research field. Q value is an important term to MPF. It determines the performance of selective frequency. The higher of Q value, the better of the selective frequency. There are many factors affecting Q value, such as the coefficients of filter and FSR. The characteristics of structure in MPF were analyzed and the effective way to obtain the high Q value was acquired.

The continuous-wave Tm: YAP laser operating in room-temperature is studied. A 795 nm laser diode pumping Tm: YAP laser crystal using the way of thermoelectric refrigeration and air cooling was applied to realize 1.99 μm thulium laser output, the maximum output power is 13.5 W, the optical conversion efficiency is 28.2% and the slope efficiency is 36%. Experimental analysis of the type of cavity affecting laser output and the operation temperature of crystal has been made.

On account of the hysteresis error of the sensor, a new kind of elasticity alloy suitable for a substrate material of the FBG sensor was studied. The elastic after effect and the elastic hysteresis effects of this alloy are very low. A new sensor structure was designed, using elastic steel and newly developed constant elasticity alloy as substrate materials and the two sensors’ pressure hysteresis errors were compared through pressure experiment. Experimental results show that the sensor in non-ferrous, made the process of pressure, and by using the new hysteresis error of pressure sensor material retardation is very small, for 0.099% error and can be neglected. Sensor has function of temperature compensation, can also measure of temperature and pressure. The sensor can be applied in oil tank, the temperature and pressure etc.

To improve the measurement accuracy of open-loop fiber gyroscopes, the effects of nonideal optical component on the measurement accuracy are investigated in detail. According to the structural feature, the Jones matrix models of different components in the fiber gyroscope are proposed and the intensity output of the fiber gyroscope is deduced with the methods of coherence matrix of partly polarized broadband source and transferring matrix theoretically. The polarization errors originating from the fluctuation of optical source power and polarization degree, the fluctuation of coupling ratio and alignment errors of the fiber fuse splice are also analyzed. Numerical results demonstrate that the fluctuation of optical source would bring the most distinguish effect on the measurement accuracy of open-loop fiber gyroscopes. The polarization errors originating from the fluctuation of alignment errors of the fiber fuse splice lay the weakest effect on the measurement accuracy.

A sort of one axle tracking Solar Photovoltaic system is presented. To solve the problem of low conversion efficiency in existing systems, the method of timing light intensity comparison is proposed to achieve automatic tracking of solar cells. Based on this, combining practical applications at present, the way of using one axle tracking can improve the Photo-electricity Changing Efficiency. Furthermore, the method of the one axle tracking is simple and firm at structure, it does not reduce the system’s resisting wind capability, so that it has better practicability and value to extend.

The Joint Fourier Transform Correlator (JTC) exists a strong and widely zero-order diffraction, which restricts the input image size and its relative position. A method is presented to place respectively the reference image and the image to be identified in the arms of Mach-Zehnder interferometer, and to make the joint power spectrum and the identified targets power spectrum subtraction. The improving method can make two original outpour items move to eh coordinate center, it not only improves the ability of identification, but also increases the input image size.

Mint image registration is an important step of the authenticity identification and defect detection of valued coins based on image recognition. On account of the specific of mint image, the texture image formed by edges is proposed to substitute the grayscale as the characteristic for image registration. The proposed registration algorithm utilizes various integral projection methods including polar coordinates projection. The translation, rotation and scale parameters are obtained by calculating the cross-correlation functions and its maximum parameters of the projection profiles of the sample image and template image. This process is iterated multiple times to achieve accurate registration. The proposed registration algorithm lays a sound foundation for further identification of authenticity and defect detection of valued coins.

Since traditional camouflage evaluation methods are restricted by a lot of conditions, objective and quantificational evaluation method that can reflect real camouflage effect is necessary. Based on the quantized color histogram, the color similarity extent between camouflage pattern and actual background is evaluated quantificationally. And the texture similarity extent between camouflage and background is evaluated by method of combining grayscale concurrence matrix and wavelet transform. The method mentioned in this thesis not only can evaluate the similarity extent of camouflage and background, but also can reflect the texture similarity extent at different observe distance.

A background matching algorithm is introduced, which used to detect and track the moving target when the video and target moved respectively. The background were aligned using relevant matching algorithm, and the moving target can be effectively extracted with this algorithm combined with inter-frame differential technique. The moving target was tacked by estimating the next position of the moving target, and matching the optical flow and the counterpart.

The remote sensing image characteristics are introduced, the remote sensing satellite and its major measuring principle are analyzed. Three correction methods have been studied for the problems of radiometric distortion, geometric distortion, and San color distortion caused by atmosphere. The prospect of the application of the remote sensing image is presented.

Aiming at the differences in texture character of different background images, a method based on GLCM and PCA for background classifying was put forward. First, 14 texture eigenvalues were extracted on the basis of the constructing becomingly GLCM, and then the PCA was executed to gain the colligation texture eigenvalue. The experimentation shows that different background images can be availably classified by the colligation texture eigenvalue.