The present status and recent advances in nanofabrication of X-ray diffractive lenses in domestic and overseas were reviewed. The research and achievements of in the past three years by authors group were introduced. For the key technology of diffractive lenses， the basic processes of electron beam lithography for large aspect ratio profiles in resists was established. By combing with Au electroplating， a solid technical background for nanoscale zone plates was successfully built up and applied to the fabrications of Siemens stars and grating based beam shapers. Furthermore， by applying Monte Carlo simulation and developing dynamics， the aspect ratio (zone height/zone width)limit by electron beam lithography was explored and the physical essence leading to the limit was discussed. A serial of diffractive lenses， such as 50—100 nm zone plates (in which the 100 nm zone plate shows its aspect ratio of 16∶1)，50—300 nm Siemens stars(in which the 300 nm Siemens star has the aspect ratio of 10∶1)and 200 nm grating based condensers(with the aspect ratio of 10∶1)were successfully fabricated. The optical characterizations of these fabricated lenses were measured in Shanghai Light Source， It demonstrates that the focusing spot of 100 nm zone plate is 234 nm. The 300 nm Siemens stars and the condensers fabricated in this work show their optical quality at the world level. The standard deviation of the illuminated intensity is measured to be 1%. Finally， this paper summarizes the development of X-ray diffraction lenses in our country in recent years， and points out that the biggest bottleneck of the development of diffraction optics is the interaction between resolution and diffraction efficiency. It puts forward some specific ways to improve the diffraction efficiency of optics and gives a technical road-map for the lens technique in China in the next five years.

For the requirements of an embedded focused imaging telescope in eXTP (enhanced X-ray Timing and Polarization satellite) for W/Si multilayers on a cylindrical mirror， W/Si multilayers were fabricated at a grazing incident angle of 0.5 ° and working range of 1—30 keV， and multilayer fabrication technologies were optimized. Firstly， separator plates and masks were mounted to collimate the sputtering particles to optimize different background pressures and working gas pressures in the deposition process and to improve the quality of periodic multilayers. Then， a novel kind of revolution speed curve was designed for controlling the thickness to make the layer thickness be uniform at the mirror axes. The separator plates were mounted on the mounting plate on both sides of the mirror to make the different axis thicknesses be equal. Finally， the depth-graded structure was designed by using the power law expression， a sample was prepared and measured in Beijing Synchrotron Radiation Facility (BSRF) and the measured results are identical with that of design theory. By optimized multilayer fabrication technologies， a multilayer with a d-spacing of 3.8 nm， thickness ratio of tungsten of 0.47 and the total number of bilayers of 80 was fabricated. The results show that the interfacial roughness of the multilayer is only 0.29 nm and the layer thickness variation on the cylindrical mirror has been controlled less than 3%. The measurement indicates that depth-graded multilayer can meet the requirements of the embedded focused imaging telescope in eXTP mission for layer quality， layer thickness uniform and energy spectral response.

The high-resolution hard X-ray (>2 keV) Fresnel zone plates with high density and high aspect ratio was fabricated for Synchrotron Radiation Light Source. The electron beam lithography and the fabrication of hard X-ray zone plates were simulated with Monte Carlo method by combining high accelerating voltage (100 kV) with Si3N4 self-standing film to reduce the backscattering. The simulation result shows that Si3N4 self-standing film substrate effectively reduces backscattering when electrons propagate in the resist， so that the structure collapse and adhesion caused by high density and high aspect ratio are overcome. By adjusting the electron beam exposure dose， hard X-ray Fresnel zone plates with the outermost ring width of 150 nm， gold absorber thickness of 1.6 μm and the aspect ratio more than 10 were fabricated on a 500 nm Si3N4 self-standing film. Meanwhile， a random support structure was introduced to realize the self support of the zone plates and to improve their stability. The focusing properties of the zone plates fabricated were tested with energy of 8 keV at 4W1A beamline of Beijing Synchrotron Radiation Facility， and a clear focusing result was obtained.

A high accuracy micro-focusing system was designed to improve the light spot focusing quality of the X-ray beam-line for Photo-emission Electron Microscope (PEEM) in Shanghai Synchrotron Radiation Facility(SSRF). The layout of PEEM beam-line of the SSRF was introduced and basic parameters of optics of the micro-focusing system were presented. Two Kirkpatrick-Baez mirrors (KB mirror) were adopted to finish the design of micro-focusing system. The design scheme of a key component-posture adjusting mechanism in the system was introduced. Namely， a three vertical linear driving equipment and a two horizontal linear driving device were combined to implement five-dimensional adjustment of the system. The working principle and process of the posture adjusting mechanism were introduced， and the overall design scheme of micro-focusing system were analyzed deeply. The mechanical performance of KB mirror system was tested， and the testing results of the horizontal adjusting mechanism and the pitch motion of first mirror were given. That the resolution and repeat accuracy of horizontal adjusting mechanism are 0.6 μm and 0.85 μm respectively， and those of pitch motion are 0.4″ and 0.5″ respectively， which are better than that of the technical requirements. The other parameters were tested as well， and the results also satisfy the technical requirements. The realization of technical targets of KB mirror system guarantees the high quality focusing of PEEM bean-line.

To improve the focusing performance of hard X-ray focusing elements， a hard X-ray Compound Kinoform Lens (CKL) based on Polymethyl methacrylate (PMMA) with a depth of 60 μm is fabricated by LIGA( Lithographie， Galvanoformung， Abformung)technology， and its excellent morphology is obtained. The main structure of CKL is a narrow pattern with a few micron widths， including the curve and the rectangular surfaces， and the width of the narrowest pattern is only 2 μm. The fabrication of the CKL is divided into three parts to ensure its excellent morphology: the transitional mask preparation， LIGA mask preparation， and hard X-ray exposure to get the finial samples. The nanopillar arrays on silicon substrate are used to solve the falling-off problem of the photoresist during the LIGA mask preparation. Then the PMMA sample with a high molecular weight is selected as the substrate to improve the rigidity of the PMMA and to relieve the adhesion and collapse of the narrow pattern in the final sample preparation. The focusing performance of CKL is tested at X-ray microscope beam-line of Beijing Synchrotron Radiation Facility (BSRF). The experiment result for an 8 keV X-ray at transmission shows its full-width-at-half-maximum (FWHM) peak size for focusing spot is 440 nm.

As the optical constant of MgF2 over-coated Al mirror effects its optical performance greatly， this paper explores the method to obtain the optical constant. Three Al/MgF2 mirror samples covered by MgF2 coatings with different thicknesses were deposited on B270 substrates at room temperature by thermal evaporation method. The layer structures of the samples such as the thickness and the roughness were characterized by the grazing incidence X-ray reflectometry. The reflectivities of the samples at a incidence angle of 5° were measured in 105—130 nm region at the National Synchrotron Radiation Laboratory(NSRL). Finally， the optical constant of MgF2 for Al/MgF2 mirror was obtained by Fresnel formulas at given wavelength and corresponding reflectivity when the coating thicknesses of MgF2， Al， and the optical constant of Al have been known. The common intersection point of the curves of three MgF2 coatings with different thickness for Al/MgF2 mirror determines the optical constant of MgF2 coating at given wavelength. The analysis and comparison show that calculated reflectivity from optical constants of this work agrees well with measured reflectivity curves in 108—128 nm region.

The microscopic response measurement of material dynamics was researched experimentally under the loading conditions of gas gun loading. Based on a flash X-ray source with the half width of 25 ns， a real-time X-ray diffraction diagnostic system was developed by using integral record devices. The measuring principle of X-ray diffraction system was introduced. Due to the duration of shocked state in the probed region and the output timing jitter of the X-ray source， the difficulty of system synchronization for X-ray pulse and shock wave in arriving probe region was discussed. Finally， the method to realize the synchronization of the X-ray pulse with the shock wave was achieved by utilizing a fine multilayered target and a piezoelectric pin with ns response time， and a real time X-ray diffraction image of LiF in the peak compressed state was obtained under gas-gun loading. Experimental results indicate that the lattice compression of LiF is 1.73% when shocked pressure is 2.33 GPa. The experimental technique provides an effective means for research of the microscopic response under a gas-gun loading experiment.

This paper focuses on the reflectivity decrease of reflective elements caused by the carbon contamination deposited on multilayers during EUV( Extreme Ultraviolet) lithography working and emphasizes the clear method of carbon contamination deposited on multilayers. The process of carbon contamination deposited on multilayers was elucidated and the damage of carbon contamination on the multilayers was introduced briefly. Several kinds of cleaning methods for carbon contamination deposited on multilayers were described in detail from cleaning mechanism， removing rate and cleaning effect， and their advantages and disadvantages were analyzed. The result indicates that the cleaning rates of plasma oxygen and activated oxygen reach 2 nm/min， but the multilayer surface is easy to be oxidized in cleaning process; and the plasma hydrogen and atomic hydrogen have relatively slow cleaning rates， they are only about 0.37 nm/min， but the multilayer surface is hardly to be oxidized. Moreover， the difficulties of different cleaning methods for X-ray multilayer mirrors in-situ are discussed.

The photon flux of the synchrotron radiation hard X-rays was measured in Beijing Synchrotron Radiation Facility (BSRF) using a free-air ionization chamber. Absolute air-kerma can be obtained with the free-air ionization chamber. The photon flux was calculated based on the relation between air-kerma and flux in the fundamental quantities and units for ionizing radiation. Because the photon flux changes over time during the measurement, an online monitoring system was applied, and the uncertainty of which is 0.03%. The air attenuation correction factor and ion recombination correction factor were measured at 15 keV synchrotron radiation X-rays. The measurement results show that the absolute air-kerma rate at 15 keV synchrotron radiation X-rays is 0.239 Gy/s, accordingly, the photon flux of the source is 5.28×109 photons/s, the relative uncertainty is 0.68%.

In order to achieve high spatial resolution and narrow spectral bandwidth imaging of laser plasma in updated SG-Ⅱ laser facility, the relations of the imaging quality of spherically bent crystal with image distance, Bragg angle and backlit size were discussed. First, spherically bent crystal imaging at diverse image distances, Bragg angles and backlit sizes were studied with a ray tracing program X-LAB. Then, a test experiment was carried out by using X-ray tube device. The results show that better two-dimension image could be obtained at the image distance of equal resolution of two dimensions, and larger the Bragg angle is, better the imaging is. Furthermore, smaller size of backlit is also helpful to improve the imaging quality. The experiment result agrees well with the expectation, a spatial resolution better than 14 μm in the image of the mesh. The results obtained contribute to the optimization design of the spherically bent crystal imaging system and precision diagnosis of laser plasma.

NiC/Ti neutron supermirror plays a key role in improving the capability of neutron optics. In order to improve the performance of the neutron supermirror, a study in microstructures and interfaces of NiC/Ti multilayers with varied NiC layer thickness was presented. The thickness, interface roughness and crystalline state of NiC/Ti multilayers were represent by Grazing incidence X-ray reflectivity and X-ray diffraction, respectively. The study results indicate that the NiC-on-Ti interface roughness was almost constant except the thin NiC layer, which has the thickness less than 2.5 nm, however the Ti-on-NiC interface roughness has relatively large changes. The NiC layers transferred from amorphous to polycrystal, as the NiC layer thickness increase from 2.5 nm to 5.5 nm, and both two interfaces and the size of Ni (111) crystalline grains almost kept constant as the NiC layer thickness continue to increase. This different variation between Ti-on-NiC and NiC-on-Ti interface roughness can be attributed to the microstructures changes resulting from varying NiC layer thickness.

In consideration of the working characteristics of a Charge Coupled Devices(CCD)， an imaging detector based on CCD and CsI scintillator was presented by using flexible exposure control and multi-frame superpositioning techniques to improve the detection efficiency of hard X-ray. An experimental platform was built， in which a nickel collimator based on a micromachining techniques was adopted as the imaging spatial target. In the experiment， the 55Fe X-ray source was adopted for direct imaging of CCD itself. Then， the 241Am X-ray source was used for both direct imaging by CCD itself and indirect imaging by the hard X-ray imaging detector. Finally， some phenomenon exhibited in the images such as drag shadows， clarity gradients， and different area brightness were analyzed. The results indicate that the techniques presented not only expands the responsive energy range， but also improves the quantum efficiency significantly as comparing with CCD itself. In addition， under exposure with 241Am X-ray sources， hard X-ray imaging with a spatial resolution better than 50 μm was achieved. The imaging detector is promising to be a position sensitive detector for future space astronomical observations.

To exactly characterize X-ray light resource parameters， a measuring method for the focal spot size and focal depth of an X-ray source was proposed by using a single-bounce ellipsoidal monocapillary X-ray condenser (SBEMXRC). The SBEMXRC was a kind of X-ray reflective imaging optics and characterized by its single total reflection imaging capability. The universal relationship among the focal spot size of the X-ray source， image size of focal spot of the X-ray source and the slope errors of the SBEMXRC was determined by the X-ray sources with known spot sizes simulated by a polycapillary X-ray lens. The focal spot size of the X-ray source with a unknown spot size was accordingly obtained by analyzing image size of the focal spot. The focal depth of the X-ray source could also be measured by the designed method. To verify the feasibility of the designed method， the spot size and the focal depth of a microfocus X-ray source in our lab were measured. The experimental results show that the arithmetic mean standard deviations of the measurement are 1.5 μm and 0.7 mm for the X-ray source with a spot size about 50 μm and a focal depth about 20 mm， respectively. The results demonstrate that the focal spot size and focal depth of the X-ray source could be simultaneously measured with the designed method. This method has potential applications in the field of X-ray sources.

To meet the requirements of the X-ray fluorescence spectrometer in Shanghai Synchrotron Radiation Facility(XAFS) for sensitivity and resolution， a synchrotron radiation X-ray fluorescence spectrometer was proposed based on three-crystal and multi-axis linkage. In experiments， the X-ray was provided by a double crystal monochromator， the dispersion structure was made of three concave crystals and these crystals formed an intersectant Rowland circle in the vertical plane to achieve the fluorescent analysis. Moreover， the dispersion structure could change in the Bragg angle ranges of 10°. Four-axis linkage of three crystal and two-axis linkage of the whole platform were implemented by high precisely controlling the driving device. Here， the displacement precision of each axis reaches 25 nm in a single step， which realizes high resolution and three-dimensional scanning work. The drivers software was programmed to improve the test sensitivity and resolution of the driver. The overall system software was designed by using the International Common Experimental Physical Control System-EPICS. By which， the spectrometer realizes the functions such as the precise control， automatic measurement， data analysis， results display and storage functions and completes a high precision and high resolution X-ray fluorescent spectrum analysis system based on Synchrotron Radiation Source. By taking the cobalt as the experimental sample， an analysis experiment was performed. The results show that the single measurement time is less than 1.5 s， the test precision is 0.4 eV and the resolution is 0.1 eV. The spectrometer can acquire and analyze the fluorescence of the sample and the operating time， precision， resolution and the repeatability are superior to those of the existing domestic and foreign equipment. It has been used in the scientific research experiments of Shanghai light source XAFS line.

The structure of a typical photoelectric tracking system was analyzed and a method to improve the resonant frequency of mechanical structure was proposed to improve the tracking speed of the photoelectric tracking system. The structure of the vertical bearing of a typical tracking frame was researched， and it pointed out that the torsional stiffness of the system was limited by the movement of a one-way thrust bearing along axial direction. Then， a structure of double thrust bearing combined with centering bearing was proposed to design of the vertical axis. By which the system stiffness was enhanced and the resonant frequency of mechanical structure was improved. A modal simulation on the improved tracking gimbal was analyzed， the resonance frequency curve of photoelectric tracking system was obtained through the experiments of vibration and sweep frequency. The results show that the resonant frequency of the system is up to 114 Hz，which provides a hardware support for the stability and higher speed of servo system. In the actual task， the maximum tracking speed and the acceleration of the system is above 150 (°)/s and greater than 240 (°)/s2， respectively， improving the tracking ability of the photoelectric tracking system significantly.

The lapping process for photonic crystal fibers (PCFs) was studied， and the end face damage of the PCFs was discussed. Based on the structure characteristics of the PCFs， a digital simulation model was established by finite element method. Through the simulation of a single grinding hole wall， the occurrence of crack damage under different cutting depths and the effect of damage caused by different grinding grain diameters on the structure of fiber hole wall were analyzed. Finally， the analysis results were verified by the PCF end face lapping experiments. The results show that FEM can effectively simulate the end face lapping process of PCF. As compared to the non hole region， the edge of hole wall is more prone to damage in the lapping process， showing a collapse area distributed along the circumference. The size of collapse area increases with the grit diameter. For no collapse area generation， the maximum cutting depth of this PCF is less than the critical cutting depth of brittle plastic transition of ordinary fiber and the abrasive paper with a thickness of 0.02 μm can be used to polish to avoid the damage on the PCF hole wall.

A novel 5-degree-of-freedom (DOF) hybrid serial-parallel manipulator was proposed. The kinematics of Parallel Mechanism (PM) of the 5-DOF hybrid manipulator was analyzed. The PM is a 3-DOF (one translational DOF and two rotational DOFs) metamorphic planar PM (2PRR)2+R with two different types of configurations: actuation redundancy and kinematic redundancy. Firstly， the kinematics model of the 3-DOF planar PM was established. Then， the stiffness matrix of the planar PM was obtained on the basis of kinematics analysis and the mass matrix was obtained by solving each equivalent component mass of the planar PM. Furthermore，the system dynamic equation was established through combining the stiffness matrix and mass matrix， and the natural frequency equation was obtained incidentally based on the dynamic equation. Finally， the stiffnesses and natural frequencies for the redundant and non-redundant planar PMs were analyzed and compared by numerical simulation. The results show that the maximum extend ranges of the stiffness around Z-axis and the first-order natural frequency reach 88.46% and 31.50%， respectively; and the minimum extend ranges of the stiffness along X-axis and the second-order natural frequency reach 52.34% and 1.90%， respectively. Thus， the stiffness and natural frequency of the planar PM have been obviously improved by adding actuation redundancy.

The valve inside a piezoelectric pump was easy to be failure when it was suffered too much stress concentration. A cymbal-shape slotted check valve was proposed to reduce the stress concentration of the piezoelectric pump. A valve-based piezoelectric pump was designed by using the cymbal-shape slotted check valve， and the working principle of the cymbal-shape slotted check valve based piezoelectric pump was introduced. The stress of the cymbal-shape slotted diaphragm was analyzed， the output performance of the piezoelectric pump was discussed and the stress of diaphragm on fluid-solid coupled interaction was calculated. Then， a prototype for the pump was fabricated. The finite element model of the cymbal-shape slotted check valve based piezoelectric pump was established and the stress of the cymbal-shaped slotted diaphragm on fluid-solid coupled interaction were calculated numerically. The calculation results indicate that the calculated value of the diaphragm is the maximum when the driving frequency is 418 Hz， and it is 81.74 MPa in the normal driving frequency range of the piezoelectric pump. Finally， the performance of the piezoelectric pump was tested， and results show that the maximum output flow of the pump and the maximum amplitude of the oscillator are in the low frequency band. The maximum flow rate is 6.6 g/min driving by 160 V (5 Hz)， and the maximum amplitude of the piezoelectric vibrator is 165.8 μm (160 V， 4 Hz). The experimental results validate the feasibility of the cymbal-shaped slotted valve based piezoelectric pump. Moreover， it concludes that the stress of the diaphragm working in high frequencies is larger than that working in low frequencies.

To meet the requirements of a large-scale space infrared camera for a lighter weight and higher rigidity framework with complex support function， a framework entirely formed by trusses was designed.Then， the feasibility and reliability of the frame were demonstrated by structure analysis， technological test and mechanical test. The framework was combined by 58 carbon fiber struts and 21 joints to form the main support structure of the camera. It could support most of optical modules in circumference distribution and could meet the weight requirement less than 55 kg. The methods and steps for gluing the subassemblies were described particularly， then the finite element analysis and mechanical experiment were used to verify that the triangle enclosed truss structure in stress state could offer an adequate safety margin even if the adhesive strength was reduced. Mechanical experiment results show that the basic frequency of the framework is 90.4 Hz， which is consistent with the results of finite element analysis and higher than the limited frequency of 60 Hz.

To alignment the posture between the camera and the star-sensor on the GF-2 satellite， a high precision automatic measurement system was established. For the measuring system， a high precision measurement algorithm based on multi-sensor fusion， an automatic measurement plane based on design data， and a cube mirror normal measurement method based on geometric figure recognition were investigated. The system was mainly constituted by a double column guideway system， a precision turntable， and an autocollimation theodolite assisted with CCD camera. The instrument posture was calculated by fusing the data of the pitch angle and the horizontal angle of the theodolite and the rotation angle of the turntable. In measurement， the system was fixed and calibrated， and an automatic measurement plan was generated firstly. Then the theodolite was driven to the desired location and desired angle. If there was no collimated light from the mirror to the theodolite， the CCD camera would search the cube mirror and calculate the deviationt of the theodolite. Then the location and angles of the theodolite were adjusted and the posture of the cube mirror of the instrument was measured. Experimental results indicate that the precision of the system is within 5″and the maximum deviation compared to the standard value is 4.1″. The system has been used in the assembly of GF-2 satellite， its maximum measuring standard deviation is 3.5″， and satisfies the system requirements for higher precisions and rapid speeds.

The fusion recognition of multi-band images can extend the application range of recognition systems. A fusion method based on convolutional neural networks (CNN) was explored and designed in this paper. Based on the AlexNet network model， it was extracted that the ship target features of three wave band images concurrently in visible light， Middle Wave Infrared (MWIR) and Long Wave Infrared (LWIR) bands. Then， it performed the feature selection for concatenated three-band eigenvectors by using the mutual information method and determines the dimensions of fusion eigenvectors according to sorting the importance of concatenated feature eigenvectors. Finally， three fusion methods named as Early fusion， Middle fusion and Late fusion were used to verify respectively the effectiveness of the proposed algorithm according to the features extracted from different levels. An available ship target dataset in three bands containing 6 categories of targets and more than 5 000 images was established for our experimental verification. The results show that the recognition rate from Middle fusion reaches 84.5%. Compared with Early Fusion and Late Fusion， it increases by 8% and 12%. Moreover， the recognition rates of all three fusion methods have been improve significantly as compared to that of the single band recognitions at the same application scene.

Because the existing motion recognition method couldnot be directly used in human micro-expression action recognition. A spatio-temporal pyramid Weber Local Descriptor (STPWLD) was constructed and an automatic recognition algorithm of human lower limb micro-expression action based on dictionary learning according to characterize human lower limb micro-expression action was designed. With the method， the features of human lower limb micro-expression action was extracted by the STPWLD. Then， the dimensions of STPWLD feature were reduced by the principal component analysis. Furthermore， the sub-dictionaries of human lower limb micro-expressions action was established and these sub-dictionaries were connected in series to construct a general action dictionary. Finally， the influence of the parameters of the algorithm on the recognition results was analyzed， and the optimal value of these parameters was determined. It shows that the optimal value of pyramid scales is 3， the optimal feature dimension of each action after dimension reduction is 30， the optimal number of atoms in each action dictionary is 40 and the optimal value of sparse threshold is 10. The experimental results indicate that the recognition rates of the proposed algorithm for 10 kinds of human lower limb micro-expression actions are all between 0.83~0.91， and the average recognition rate is 0.86， higher than that of other algorithms. The algorithm is suitable for the classification of human lower limb micro-expression actions and improves the classification recognition rate effectively.

To estimate quickly the relative pose of a space target based on point cloud， a Binary Rotational Projection Histogram (BRoPH) feature descriptor was proposed. Firstly， a Local Reference Frame (LRF) for the feature point was established; Then， the density and depth images were generated under different views by rotationally projecting the local surface of feature point， Finally， the multi-scale binary string of the feature point was produced based on the images. For implementing the pose estimation of space target in real time， a Hamming distance threshold based feature matching strategy was proposed further to exclude false matching pairs to accelerate coarse pose estimation procedure. The comparison experiments were performed with SHOT descriptor and FPFH descriptor. The results demonstrate that BRoPH achieves an accurate pose estimation with only about 1/80 average memory cost of SHOT and FPFH descriptors. The average attitude error of BRoPH is under 0.1 °， and its average translation error is less than 1/180R. Besides， the Hamming distance threshold based feature matching strategy speeds up the subsequent RANSAC by 7 times， and the overall pose estimation frequency exceeds 7 Hz， which is 3 to 6.8 times faster than those of SHOT and FPFH descriptors respectively. The proposed feature descriptor is compact and efficient， and the pose estimation method is accurate and robust for the requirements of space target pose estimation.

Current watermarking algorithms use mostly pseudo random sequences to generate watermarks. These algorithms emphasize the improvement of imperceptibility and robustness of the watermarking but ignores its safety. To improve the watermarking safety， a new secure communication oriented airspace watermarking algorithm with a tamper locating function was proposed based on the true randomness of quantum key. According to the principle of BB84 protocol， the quantum key with real randomness and absolute security generated by quantum key distribution mechanism was used to generate binary image watermarks. Then， in combining with a quantum key matrix model and 8-neighborhood random location idea， the quantum watermarks were embed into a carrier image dynamically and randomly. At the end of the transmission， the receiver could quickly extract watermark information， accurately judge the integrity of the watermark image and the security of the transmission process， and could perform tamper localization. The concealment， robustness and tamper location of the algorithm were tested. The experimental results show that the algorithm is simple， safe， and has higher safety and covert and its tamper locating accuracy is 3 pixel×3 pixel. The algorithm can be widely used in the secure transmission of digital images.