The image matching based on features is one of the important approachs. A lot of works have expressed that feature based on matching takes powerful advantage for urban aerial image which is one of the most difficult image to deal with. A construction of a new object function is given. A novel matching method for aerial imagery of urban area based on measurement feature is proposed. The experimental results show that the proposed approach is simple and effective.

The Hausdorff distance between planar sets of points is known as a good method to compare binary images. We present an algorithm to match terrain using the modified Hausdorff distance as the measure of the difference between two images. The terrain structural lines is obtained;then the Hausdorff distances are computed;finally the Husdorff distances is compared. The minimal Hausdorff distance means the best matching result. The experimental result shows that the proposed method is effective for matching terrain.

A common problem about matching is how to select the best subset according to a given criterion. The key factors of subset selection are uniqueness and lack of ambiguity. Uniqueness could be decided by the number and magnitudes of the peaks in the cross correlation function of the original and subset functions. Lack of ambiguity could be measured by the correlation length or 50 percent width of the main correlation peak. The smaller the correlation length, the greater the probability of detecting the correct matching position. In this paper, a method is presented to select the best subset according to the minimum correlation length. The analysis can be extended to the two , three and higher dimensional problems. In discrete case the problems can be solved by using the fast Fourier transform (FFT) algorithm.

Calculation of confidence level for image matching is introduced, which is based on the assumption that image pixels are independent each other and no spatial correlation is in the scene. A large number of tests on the real sensed image has been implemented to estimate the accuracy of calculation. It shows that the value of the confidence level can be used to judge the credibility of matching result, especially when the confidence level is close to 1.0 or less than 0.5, however the matching position is uncertainty when the confidence level is higher than 0.5 and less than 0.99. Further more, how to improve the estimation accuracy of the confidence level is discussed.

Imaging characteristic difference between infrared target and decoy at the stage of extended target is studied. Method of discriminating target and decoy by the difference of global movement pattern is bring forward. To qualitatively deduce global movement characteristic such as acceleration, rotation and rolling by the variation of object′s global shape feature in image sequence is used as the discrimination feature to identify decoy from true target. Simulation result is given.

The present development and research of the imaging infrared system performance model at abroad and domestic are reviewed. On the basis of the analysis of static performance and dynamic performance model, the basic theory of two kinds of performance model is given respectively, then the advantage and disadvantage of several kinds of static performance models are described. Three kinds of module of the dynamic performance model are discussed. At the same time, the field applications of the performance model are introduced. The developing trends are prospected.

The algorithms for iris recognition set forth by Daugman can perform identification accurately and quickly, but the implementation about the algorithms has not been reported in open literatures. The algorithms are studied and implemented. A new method combining coarse location with fine location is proposed to locate the iris quickly. Furthermore, only the real parts of filters are used during the filtering, which can decrease code length and still have good results. The detailed procedure of implementing the algorithms is given, including preprocessing, constructing Gabor filters and the computation of iris codes and Hamming distances. The results show that this approach has very high speed of operation and good results of feature extraction and can be used for actual personal identification system.

Detection of low signal noise ratio small even point moving target in the condition of complicated background is discussed. A new approach based on Kalman filter, mathematical morphology, motion analysis and two stage hypothesis testing to detect and track point target in such condition is presented. The experiment results with real IR image sequence show that the proposed method can greatly improve the SNR, and effectively detect and track the moving point targets. It can especially dissolve the problems that the targets miss and new targets occur during the detecting and tracking.

The objective of automatic vehicle number plate recognition systems (AVNPRS) is to accurately extract and recognize the vehicle number plate from the complicated background. AVNPRS we developed can realize functions of real time image capturing, number plate segmentation, number plate pre processing and character recognition etc. The experiment results show that the total recognition success rate can achieve 90% with relatively short time cost, about 1.5 (with a total number of 311 experimental images which resolution is 768×574).

InSb infrared focal plane array assembly, which composed of PV InSb array, Si CMOS readout, micro dewar and micro cooler, has been developed. The FPA assembly achieves a high performance with mean detectivity 1×10 11 cmHz 1/2 W -1 , quantum efficiency ～72%, non uniformity ～16%, and operability ～99.5%. The assembly specifications meet the requirement for missile guidance, and out door staring imaging detection has been demonstrated successfully.

Three kinds of readout integrated circuit for 64×64 InSb IR focal plane array have been designed and fabricated. The circuit ISM64×64-2 works in direct injection and line integration mode. ISM64×64-3 works in the same mode as ISM64×64-2, but was added one sample and hold circuit. ISM64×64-4 works in direct injection and frame integration mode, so the signal to noise was increased greatly. The scheme, operating principle, peculiarities of the layouts, fabrication process, testing methods and performance of the circuits were presented in this paper.

Aluminum flyer of 5.5μm thickness has been driven successfully by mini size KIDS laser in the energy range of 100～300mJ. The highest velocity of flyer has been up to 6.6km/s, and the relationship of flyer velocity and energy density has been obtained in experiments. The setup of experiments, measurement system, and experimental technique are introduced in detail, and analysis are given on experimental measurement system. The analysis to recovery samples show that the plane character, integrity character and repeated character of flyer are very good. The whole process is solved by using Gurney theory, and academic results are in good accord with experimental results. The analysis on the application possibility of laser driven flyer in simulating aerospace particles are done. The results show that laser driven flyer is very effective in simulating aerospace hypervelocity particles.

Simulation in infrared imaging system (IRIS) with PCs does a great help for overall optimization. It is also a useful tool for system evaluation. Firstly, some adopted methods and important ideas are introduced, then some typical models are introduced and compared. Finally, some opinions are proposed.

The realized method of collimated coherent light beam for a novel robot vision system is presented. The properties of astigmatism ellipse Gauss beams of the semiconductor laser are analyzed. An optical structure of collimating system is designed. The collimation degree is tested by facula method. The experiments show the collimating device is suitable for the novel robot vision system.

According to the characteristic of high techniced air attack, the intimidation for important traffic targets and the austere challenge of our system for traffic safeguard and air defence are analyzed. It is necessary to establish integrative camouflage and defence for important traffic targets, put forward the basic structure designed for integrative camouflage and defence system of important traffic targets, and explained the validity of main camouflage and defence measures briefly and specifically, then proved that integrative camouflage and defence system is viable and easy to achieve on technology.

Integrated optical devices have been used in many areas. For all optical devices, a great challenge is the coupling between fibers and devices, because the alignment tolerances required for coupling is stringent. This is an area that has received much attention. The most common coupling method is to use precision etched Si V groove as the fixture of tail fiber of integrated optical devices. The quality of V groove is essential to the success of the coupling of tail fiber. The theory, design and process about the fabrication of Si V groove is discussed.

Software implementation of new DVB test system is presented. The main task of the software is to test and diagnose DVB system. It also can analyze the video program source of DVB. Firstly, with the using of the multi thread scheduling algorithm in the structure, it can carry out simple real time analysis of DVB stream which is the key to realize on line system test. Then, it can be used to deal with the more complex and sophisticated off line DVB stream analysis. The analysis results of this software include information about SI tables, PCR, as well as errors detected in DVB stream.

We precisely expose a problem occurred during re quantization process in MPEG 2 transcoding. Firstly, we expatiate the origin of the problem;then, we systematically propose a scheme which uses MSE rule to reduce the re quantization errors. In the end of this paper, we formulize the differences between traditionally used method and our proposed method. Theories and simulation results show that our proposed scheme efficiently reduces the re quantization errors while results in higher picture quality.

Natural video coding in MPEG 4 is concerned. The syntax structure of MPEG 4 video is illuminated followed by the general description of the video coding model. The key technology involved in the MPEG 4 video coding is analyzed in detail as follows: VOP generation;coding of binary and gray level α plane;motion estimation, motion compensation and motion prediction;texture coding;object based scalability include temporal scalability and spatial scalability;resynchronization, error concealment and error refreshment method provided by MPEG 4;sprite coding;zero tree wavelets based coding for still picture. The differences with other video coding standards such as MPEG 1, MPEG 2, H.263 are pointed out, which make the three distinct features of MPEG 4 stand out. The video application over the network based on MPEG 4 is shown, and the test results are given and analyzed.

The stability of feature vectors based on SVD decomposition is very low for the variation of target′s size and orientation. On this problem mentioned above, a new target feature extracting method is presented. It is a matrix describing method based on scale transformation. These matrixes of all same kind of target samples getting through transformation mentioned above have the similar energy. The matrix method proposed extracting target′s features from SVD feature values can overcome shortcomings of extracting SVD feature directly from matrix of the images. The experimental result shows this method has favorable stability to target matrix′s SVD feature vectors, and high invariant property to displacement, rotation and scale. It is a practical feature extracting method.

The local geometry property of gray image is studied, and the local region of an image in which pixels had similar or same gray value was took as an integer, and the aim of image compression was achieved. Computer simulation experiments showed that this method could obtain higher CR (compression rate) and higher PSNR (peak signal noise rate), remove the block effect satisfactorily, keep target details effectively, and had higher practical value.