In wheel set online detection based on structured light projection image technique, the wheel set light-section curves acquired by CCD are heavily distorted. The camera calibration method and calibration precision will directly affect the wheel set online detection precision. According to the nonlinear distortion imaging model, Tsai non-coplanar calibration algorithm based on a moving planar calibration plate was proposed. According to the radial alignment constraint and first-order radial distortion formula, the least square method and Levenberg-Marquardt optimization algorithm were used to calibrate the internal and external parameters of camera. The calibration plate and its three installation location were designed, on the basis of the requirements of on-site installation position and the key point regional distribution of the light-section curves. The extraction of image feature points and the calibration of camera parameters were realized. The experimental results show that the calibration precision of Tsai non-coplanar calibration algorithm was less than 0.1 mm. The calibration precision meets the demands of wheel set online detection.

In order to accurately measure the sight axis of the single axis photoelectric measuring system, based on the invariability of the point on principal optical axis of linear CCD imaging position when the object distance varies, a measurement method of sight axis on cooperative target is presented, and its theoretical derivation is done. Fix the cooperative target in a position, and place the uniaxial optical measuring system on a movable horizontal platform. While mobile the uniaxial optical measurement system follows the slideway, it will only change the distance between the optical measuring system to cooperative target, without changing the height between the axis of optical and cooperation target. Finally, conduct an experiment use of a practical photoelectric measuring system and cooperation objectives. The experimental results show that the accuracy of the measurement method can achieve -0.84".

In order to overcome the difficulty during measuring the external parameters of the camera with a long object distance, a calibration target bar was used as a medium. Its precision can be verified by the binocular vision measurement. By means of analyses of the influence of the cooperative targets’ coordinate in the world coordinate system on the internal and external parameters of the camera, the influence of the measured cooperative targets’ coordinate by the binocular vision measurement can be determined. The result shows that the geodetic coordinate error of the cooperative targets in the calibration target bar can affect the measurement error of the measured cooperative targets. The result shows that, for the propose that the measurement precision requirements of the cooperative targets should reach 10 mm, the X axis precision requirement should be 0.621 9 mm, the Y axis is 13.341 2 mm, and the Z axis is 0.729 1 mm.

A calibration method is proposed based on the error analysis of the three links prism simultaneous polarization measurement system. The polarization direction of polarizers is determined by the curve fitting method, which makes the error of the polarization direction for each polarizer less than 0.5 degree. The inconsistency of intensity responses of three channels is solved by establishing and index table of intensity response of different exposure time and aperture. Additionally, we apply Harris corner detector to compute pixel offset and the effective pixel range in the three channels of polarization image. Experimental results show that after the calibration, the measurement error of degree of polarization is less than 5%, and the measurement error of angle of polarization is less than 3％.

In order to improve the accuracy and efficiency of existing industrial laminated thin steel-plates counting method, a thin steel-plates counting method based on Array CCD image detection was studied. An OTSU improved algorithm within rough background separation and self-adaptive segmenting threshold corrected was proposed, which was based on CLAHE (Contrast Limited Adaptive Histogram Equalization) principle, dimensional mean-filtering and morphology optimization principle of expansion and corrosion. Due to the shortcomings on missing or error peaks detecting in parameter space of traditional Hough transform, an algorithm based on auto-correlation principle and feature segmentation was proposed. Experimental results demonstrate that the proposed counting method for stacked thin steel-plates of specification 0.5 mm above is real-time, and has high statistical accuracy and good robustness.

A realization of the multi-frequency technique for Kelvin Probe Force Microscopy (KPFM) is studied. In the multi-frequency KPFM, the first eigenmode of the cantilever probe is used for topographic imaging, while the second one is resonantly excited for surface potential. Thus, an additional feedback is needed to adjust the probe automatically. An analog feedback controller with bandwidth of about 5 kHz is designed based on the principle of conventional Proportional-integral Controller (PIC). Employing the controller, a novel multi-frequency KPFM is developed, whosesensitivity for surface potential measurement is demonstrated to be better than 5 mV. Topography and surface potential distribution for a kind of dielectric film with the injection of charge are imaged with the multi-frequency KPFM, suggesting its application in the characterization of electrical material and devices.

Micro displacement measuring device with high precision based on the principle of grating interference has high voltage sensitivity, thus weak jitter can cause irreparable impact on measuring results. These jitters include the air jitter, the ground vibration, and the fluctuation of the light source, etc. It becomes crucial for how to restrain the influence of outside jitter on the measurements of micro displacement effectively. Firstly, a series of experiments and analyses was carried on for the power fluctuation of the linearly polarized He-Ne laser, which drew that the power fluctuation range of light was greater than 7% which exceeded the fluctuation limit of 1%. Then, using the compensation measures of two signal acquisition and scaling operation, anti light disturbance came true. The analysis of experimental data shows that the effect of light source fluctuation on output voltage can be suppressed effectively through the compensation of optical path and scaling operation. The final amplitude of fluctuation of the output voltage is reduced by 60% than the initial state, and the displacement measurement error reduced from 1.95×10-7 m to 1.40×10-8 m.

In order to obtain the transient lubricating film thickness of slider bearings, a conformal contact film thickness measurement system has been developed based on dichromatic interferometry. Firstly, the basic setup of the slider bearing tester is introduced together with its working principle. The method and the resolution of dichromatic equivalent wavelength for relative large film thickness measurement are also thoroughly analyzed. Then, a plain and feasible approach is proposed for transient film thickness measurement, whereby fringe order of equivalent wavelength is conducted through dark characterized fringes and the corresponding fringe order of red or green monochromatic light is identified with modulated intensity signal to infer the film thickness. Theoretical and experimental results indicate that the well built system can be used to measure transient lubricating film thickness. The resolution of the dichromatic interference method is 0.6 nm with a measurement range from 10 nm to 5.5 μm, and the precision is 22 nm.

The key to the sensing application of the FBG is the demodulation of the reflectance spectrum, and OLCR technology is an important method to obtain the characteristic of FBG reflectance spectrum. Based on Michelson interference optical path, a set of OLCR-FBG detection system has been established, hence, the operating principle and implementation mode of the system are expounded. In the study, we propose that the pivotal issues in the system realization are the acquisition of weak interference signal, uniform sampling and discrete data processing. In the experiments, the interference signals of balance detection and unbalance detection, together with the signals of general translation stage and air-bone translation stage are detected. Results show that the SNR of interference signal is improved by balance detection structure; Low-noise air-bone translation stage makes great contribution to uniform sampling by avoiding the reflectance spectrum signals being covered by the noise; Through data processing, FBG reflectance spectrum can be quickly demodulated from interference signal, and repeatability of the Bragg wavelength is 4 pm.

For the problem of being susceptible to the influence of noise, low accuracy in micro in-plane displacement measurement using traditional digital speckle correlation in frequency domain, digital higher-order vortex speckle correlation method in frequency domain is proposed. First, it is found that higher-order vortex speckle pattern has an ability to resist noise and the output surface of correlator can get smaller and brighter correlation points according to the experiment of generating higher-order vortex speckle pattern, then using higher-order speckle pattern as measured object. Second, theoretical calculations are deduced, and then new patterns before and after displacement of the higher-order vortex speckle are synthesized using subtract operations. The two correlation points can be gotten on the output of correlator, which changes the current status that only has one correlation point using traditional method, effectively counteracting background noise. Finally, we design a new filter to generate the relevant spot diagram, whose correlation spot map peaks become sharp. Experiment results show that the algorithm has made great advance in reducing noise and improving accuracy, which can maintain relative error less than 2% under different noise.vortex; matching filter

In the framework of traditional particle filter, object is always drifted for the illumination light changing, deformation and poses changing, which affects the accuracy and robust performance of tracker. The reason is that there is not smooth in the objective function. Error will accumulate during the state updating process. A new stack particle filter is proposed to represent target model by the kernel histogram in the discrete fields. It can track the target from coarse to fine by the modified particle filter. Extensive experimental results show that the proposed algorithm performs 20% improvement in the accuracy, favorably time expedition and robustness against state-of-the-art methods.

For real-time accurate target segmentation is a major factor restricting infrared radiation characteristic measuring real-time processing system, a real-time accurate segmentation method is proposed based on miss distance of target. Firstly, this article analyzes the characteristics of the target and common image segmentation algorithms. Then, choose One-dimensional Minimum Error method (OME) and Two-dimensional Minimum Error method (TME) as the main study object for their suitable target characteristics and good segmentation results. Finally, propose using the miss distance to reduce gray level and processing area after analyzing the bottleneck of OME and TME, realized real-time accurate segmentation. Simulation results show that a frame processing time is less than 10 ms, which can meet the requirements of the infrared radiation characteristic measuring real-time processing.

In order to detect small target among dense star background, an improved difference target detection method is proposed. Firstly, the registration between two consequence frames is acquired by use information of telescope’s encoder, position and direction. Then previous frame is dilated when the difference, which can reduce error caused by atmospheric disturbances and registration rounding. Secondly, after suppression of background star edge as much as possible, candidate target is obtained by segmenting, target clustering, and labeling of differenced image. Residual background star edges are removed by using the position of candidate targets to remove star edges which not get rid of in previous procedures. Finally, double stars are removed by information of the amount of registration. After a series of pretreatment, dense background star can be removed efficiently by this method. Experimental results show that candidate targets could be kept within 20 by this proposed algorithm, for 2 k×2 k, clustering star background image.

A new method and thinking is put forward according to traffic signs. The saliency maps are respectively calculated in L, a, b color channel and H, S, V color channel of two kinds of color space Lab and HSV. And through the number of pixels of the saliency maps obtained in high, median, low brightness range to screen which saliency map in Lab and HSV color channels is effective. Finally, the final saliency map is obtained through the fusion. The experiments show that the method is easy to be implemented and can detect salient region of traffic signs fast and effectively.

Size of laser facular control unit in laser target simulation system for guided munitions was studied. According to the design index that this system requires for variation range, the maximum rate of changes and control accuracy of the field angle of facular, mathematical model of size of laser facular control unit was established. Then, servo control system with unidirectional limited angular position was composed of an ELMO driver-controller and a DC torque motor，as a result, current feedback, velocity feedback and position feedback made the closed loop control be realized inside the system. Experimental results of the system testing indicate that closed loop servo bandwidth of the DC torque motor is 82.1 Hz，and field angle of facular changed from 0.5 mrad to 5 mrad is 112.1 μrad, which is less than 0.15mrad that the design requires. The maximum error of the field angle of facular changed from 5 mrad to 90 mrad is 326.1 μrad, which is also less than 0.5 mrad that the design requires. Within the variation range of the field angle of facular, its error is in the designed reasonable error range, which can ensure that the control strategy meets the performance requirement of system.