Terahertz radiation (0.1 THz~10 THz) has attracted extensive attention of researchers recently, because of its prominent detecting ability and its noninvasive and non-ionization properties. Terahertz technologies can be categorized into terahertz imaging and terahertz spectroscopy according to the manner of detection and signal processing. With the rapid development of terahertz technology, detection of macromolecule and imaging of tissues have achieved impressive progresses. Differences of water content and variations of structure or component are essential mechanisms of terahertz biomedical imaging, which was used in identifying different biomedical tissues. Terahertz spectroscopy is an edge technology for recognizing biomolecules, cells and tissues, based on their individual terahertz spectral fingerprints by assessing their absorbance, reflective and refractive index. Based on its properties, terahertz technology has great potential ability for clinical application, especially for real-time and label- free identification. However, this technology needs to overcome several difficulties, like biological safety. In this review, we introduce the applications of terahertz imaging and spectroscopy in medical science and medical research progress, and also discuss the difficulties of terahertz technology and potential biological safety.

For the problem of constant false alarm rate (CFAR) detection in Weibull clutter background, a CFAR detector—cycle elimination TLME-CFAR detector is proposed. The detector calculates its detection threshold through the estimation of two parameters ofWeibull distribution, which is based on TL-moment estimation. The effect of the interference target and the strong scattering point are then eliminated by the cyclic elimination method. This paper proved that the proposed detector is a CFAR detector, and then the performance of the detector is studied by Monte Carlo simulation and compared with the MLH-CFAR detector. The result shows that the cyclic elimination TLM-CFAR detector has very nearly the same performance with MLH-CFAR detector. The detector avoids iterated operation of maximum likelihood estimation, and improves the efficiency and applicability of detection algorithm.

In order to obtain photonic crystal fiber with high birefringence and flattened dispersion, we propose a new photonic crystal fiber structure with an elliptical air hole as the core surrounded by square air holes. In this paper, the effects of different fiber core ellipticity and different filled material on birefringence, dispersion and nonlinearity of photonic crystal fiber are discussed. The results show that at the wavelength of 1.55 μm, when the core ellipticity of the filled material is the same, the maximum birefringence value is 0.37 and the maximum value of nonlinear system is 277.76 W-1×km-1. When the fiber is filled with different materials, the maximum birefringence value is 0.34 and maximum nonlinear value is 307 W-1×km-1. In addition, in the wavelength range of 1.26 μm～1.8 μm, the nearly zerodispersion flattened characteristics are achieved. The range of variations is no more than ±12.5 ps/(nm×km), and the bandwidth is 0.6 μm.

The sparse subaperture stitching, the accuracy of which is closely related to the arrangement, number and size of subapertures, is one of the main methods of quality testing for large aperture optical systems. A mathematical model was established to deduce the relation curve between the subaperture number k and fill factor M when the value of k ranges from one to infinity. As a result, the optimal arrangement layout, consisting of seven sparse subapertures, was obtained for the detection systems below 1.5 m. Autocollimation interference detection of Φ200 mm validated the rationality of the arrangement.

In order to deal with complex scene change problem in the tracking process, we propose a tracking algorithm via multiple feature fusion. Under the framework of particle filter, dynamic feature weights are calculated by making an uncertain measure of each feature in the tracking process, which results in adaptive feature fusion. The algorithm uses the complementarity of color, space and texture features to improve the tracking performance. Experimental results show that the algorithm can adapt to complex scene changes such as scale, rotation and motion blur. Compared with traditional algorithms, the proposed algorithm has obvious advantages to complete the tracking task.

The mirror assembly of an electro-optical tracking and pointing system for a space borne laser communication system is studied, three flexible supports are contrasted, according to practical applications of space load, the structural stiffness advantage and surface figure of the three flexible support scheme have been evaluated. The analysis results show that the surface figure RMS of neck side grooving flexible support scheme resisting micro- gravity and thermal environment change can reach 2.05 nm and 8.88 nm, the fundamental frequency mode is 926.1 Hz, in the balance between the surface figure RMS and the higher stiffness resisting hevibration damage, the flexible design is most reasonable. On this basis, the parameterized design of the flexible support structure of the reflector is completed and dynamic analysis have been done. The maximum stress of the frequency response is 96 Mpa, which is less than the material’s tensile strength limit. The results of random vibration analysis show that root mean square of acceleration response is 11.14 g RMS, meeting 3σ law. Finally, a 0.2 g sine sweeptest proves that the relative error of the modal analysis is 2%, the experimental result show that the analysis results are basically accurate and reliable, that is, flexible support design is reliable to meet the requirements of use.

The installation of binocular vision at the end of a manipulator reduces its availability in environments with obstacles. To deal with the problem, this study puts forward a target localization method using a laser and the monocular hand-eye vision. In the proposed method, the centre of the laser spot is obtained by the hand-eye vision, and the geometric relations among the laser emission point, light-spot and the optical axis of the camera are used to calculate the distance. Then, the D-H method is employed to construct the coordinate conversion system, so that the location of the target can be calculated. The measuring precision is negatively correlated with the distance, and it is suitable for the measurement in medium or short distance. Compared with the commonly used binocular measurement methods, the proposed method uses fewer cameras, which reduces the width of the measurement system on manipulators, and makes it more applicable to narrow workspace. Moreover, it also improves the effective load capacity of manipulators.

The wheeled polishing technique based on industrial robot is established by combining the advantages of robot control and wheeled polishing technology. The feasibility of wheeled polishing tool in high-precision polishing processing is validated by numerical simulation. The wheeled polishing tool installed at the end of the robot is designed, and the robot wheeled polishing control logic and framework are analyzed, thus a robot polishing control mode based on trajectory and dwell time is established. By carrying out the experiments of robot’s single-point and belt removal characteristics, the parameters of robot wheeled polishing processing are confirmed. At last, the automatic polishing processing of the mirror surface is done. The surface error decreases from the initial value of components PV: 2.357λ (RMS: 0.565λ) to PV: 1.431λ (RMS: 0.242λ) as expected. As the research shows, the industrial robot with wheeled polishing tool is an effective surface polishing method, which has great potential in high-precision aspherical mirrors.

In photovoltaic systems, the output power curve of solar battery has multiple peaks, under the partially shaded condition. Traditional maximum power point tracking (MPPT) search method often traps in local extremum, which causes the loss of the global maximum power point even generates oscillation and leads to instability of output. An improved bat algorithm (IBA) is proposed and used to find global optimal point, by introducing chaos search strategy in initial arrangement which can improve the uniformity and ergodicity. The self-adapting weight is introduced to enhance the global searching ability of previous processing and the local searching ability of late processing, and Levy flight is introduced in the same time to create the saltation velocity to jump out the local extremum. Dynamic contraction is also used to decrease the search section more effectively, so as to avoid premature convergence of the population affected by the local extremum. The simulation shows that modified bat algorithm can find the global optimal point fast, with high precision, under the partially shaded condition.

Our country is at a developing stage in the area of high-frequency observation of oceans. As an important approach of observation of ocean, the development of the geostationary ocean radiometer is of great significance. In this paper, we analyze the image rotation errors introduced by the optical system of plane imaging with two-dimensional pointing reflection mirror, propose the image rotation elimination method for two-dimensional scanning and the verification method. It is proved that the proposed rotation elimination method can reduce 39% of image rotation. This indicates that the algorithm greatly improve the accuracy of two-dimensional directional imaging and improve the accuracy of remote sensing instruments. Therefore, the work lay a good foundation for image processing in two-dimensional pointing plane imaging system.