Differential absorption lidar is an indispensable tool to measure atmospheric CO2 for temporal and spatial distribution. The 2.0 μm wavelength Ho:Tm:YLF/Ho:Tm:LuLiF lidar were used for remote sensing atmospheric CO2 with heterodyne receiver. The KTP parametric oscillator was injected into the seeds, and a differential absorption lidar (DIAL) system of 1.6 μm by using the photon counting technique to profile the atmospheric CO2 in the troposphere under 7 km. Modulation continuous wave seed laser intensity and the integral path differential absorption lidar, common 1.6 μm optical fiber amplifier, and the use of correlation detection technique, have unique advantages and characteristics in the detection of atmospheric CO2 column concentrations. The space program for detecting atmospheric CO2 column concentrations is NASA's ASENDS (active sensing of CO2 emission over nights, days, and seasons) mission which adopts the method of pulse and integral path differential absorption. The absorption of CO2 gas cell as a reference to stabilize seed light frequency and the precise control of cavity length lock the on-line wavelength of transmitter, is the key technique for DIAL to measure atmospheric CO2.

Deep convolution neural network has demonstrated excellent performance in target detection and recognition tasks. However, few training samples and optimization design of deep models are two main problems to be solved when applied to SAR target recognition. This paper proposes an algorithm for SAR target recognition by combination of two dimensional random convolution features and ensemble extreme learning machines. Firstly, two dimensional random convolution kernels with different widths are generated, and convolution and pooling operations are performed in input image to extract random convolution feature vectors. Secondly, random samplings based on ensemble learning theory are done for extracted feature vectors to improve generalization performance of classifier and reduce training time, and base classifiers are then trained by extreme learning machines (ELM). Finally, majority vote method is adopted to combine the classification results of base classifiers. SAR target recognition experiments based on MSTAR database were performed, and experimental results demonstrate that, training time has dropped by ten times due to fast training capability of ELM, and the proposed algorithm achieves comparable classification performance with deep-learning-based methods which use data augmentation and multiple convolution layers. The proposed algorithm has the advantages of easy implementation and fewer adjustable parameters, and improves classifier’ s generalization performance through adoption of ensemble learning ideas.

Since the coherent noise affected the quality of the Fizeau’s interferograms in the large aperture, the coherence of the beam was changed by rotated diffuser to reduce the noise of the interfering system. The relationships among the speed of the rotated diffuser, the contrast of the fringes and the SNR of the system were simulated. Then, the control parameters of rotated diffuser would be required in the optimum interference fringe. The interference images were obtained under different control parameters, and the fringe contrast and system SNR of each image were analyzed. The results showed that the contrast can be reduced by increasing the speed of the rotated diffuser in a certain extent, but the SNR can be improved effectively and it was convenient to process the interference image later.

In high-precision fiber-optic gyroscope (FOG) system, the spike noise of DC-DC power source can lead to a considerable disturbance to the signal processing circuit of FOG, which results in a sampling error. In this work, the cause of spike noise and the influence mechanism were clarified. The slew rate control technology was researched and proved to be an effective solution to prevent spike noise of FOG power source. Using slew rate control technology, a kind of low-noise power module has been developed and applied successfully in the FOG system. This power module consists of DC-DC circuit and LDO circuit, and slew rate control circuit was used in the DC-DC circuit to realize low-noise performance. The peak-to-peak noise value of the developed power module was tested to be about 1 mV in a bandwidth of 200 MHz. Two typical FOG systems were tested with the use of this lownoise power source, and their output noise improvement were 3.1% and 4.4%.

In order to obtain the change of posture of moving objects in wind tunnel experiment, a method of single- camera visual pose measurement based on three-dimensional topography model of object surface is proposed. The method uses the multi-perspective imaging principle to solve the target pose, obtains the feature corner point on the target as the characteristic point needed for the solution, and proposes to use target’ s 3D surface topography information to obtain the geometric relationships between feature points. In this paper, the accuracy of the measurement method is verified under the laboratory conditions. The average accuracy of displacement is 0.23 mm and the mean square error is 0.234 mm. The accuracy of the pitch angle, yaw angle and roll angle are 0.08° , 0.1°and 0.09° , respectively, and the mean square error are 0.485° , 0.312° and 0.442° . The experimental results show that the method can be used for practical measurement.

The locomotive running gear 3D point cloud data are obtained by line-structured laser scanner, and the bolts on the locomotive running gear under the 3D point cloud data are recognized and located automatically. Firstly, fast point feature histograms (FPFHs) of the key points are calculated to describe the 3D features, and the target region is matched with the preselected bolt template. Then, K-means clustering is carried out on the weighted match point set using uniform seed points. Finally, the Hough transform method is used to establish a strict classifier for the clusters, and the existence and precise position of the bolts are determined. The experimental results verify the effectiveness of the proposed method.

In order to measure the instantaneous wavefront of large aperture optical elements, a method based on the structure of oblique incidence of reflective shearing point diffraction interferometer is proposed. A lateral displacement between the reference wavefront and the test wavefront is formed after passing this structure. The shear of two beams introduces linear spatial carrier frequency to the point diffraction interferogram. After receiving a good contrast interferogram, wavefront phase is retrieved by Fourier transform ( FT) automatically to realize the dynamic measurement of instantaneous wavefront. The optical path is up to 20 m, so the air current is a significance factor to the result. Besides, because of the air current, the system itself can be seen as a instantaneous wavefront happening and measurement of large aperture optical elements. The results indicate that the root mean square value is in accord with that acquired by SID4 wavefront sensor (less than 1/50λ),so about the repeated accuracy. The method proposed can be applied in high resolution and accuracy measurement of instantaneous wavefront.

Super-resolution reconstruction plays an important role in reconstructing image detail and improving image visual effects. A new effective super-resolution method is proposed. Firstly, we extract the geometric features of the image patch to construct the decision tree, which will be used in patch classification in a supervised way. Then, we train the high-resolution and low-resolution dictionaries based on K-SVD independently for different types of training sets. Finally, we solve the mapping matrix for the coefficients between the high-resolution and low-resolution training set, which are used to map the low-resolution coefficients to high-resolution coefficients during the reconstruction phase to ensure accurate and fast reconstruction of the image patches. The experimental results show that the proposed method has a significant improvement in the reconstruction effect compared with other classic methods.

In order to overcome the problem that white balance failure caused by white region detection error in automatic white balance, this paper proposes a white balance method based on dark channel prior. First, get the dark channel image, then extract the white region in the image according to the dark channel, and then remove the region with high saturation. Finally, in order to correct the color and ensure that the image brightness does not change, we calculate the correction gain in the CIE-XYZ color space relative to the luminance channel Y. Experimental results show that our algorithm has achieved good results both in subjective and objective evaluation compared with some classical algorithms, and the rate is greater than 150 frames/s on embedded devices.

In view of the drastic increase of storage resources and transmission bandwidth requirement for high dynamic range (HDR) video compared to the traditional low dynamic range (LDR) video, we propose a dynamic rate distortion optimization algorithm based on visual perception for HDR Video encoding to improve the performance of high efficiency video coding (HEVC) Main 10 for coding HDR video. With the information of visual selective attention, we design a non-uniform distortion weight distribution strategy to different regions of interest and improve the conventional method of distortion calculation. At the same time, in order to further eliminate the perceptive redundancy in HDR video coding, the texture characteristics of video content are used to adjust Lagrange multipliers adaptively, which is applied to the encoder to dynamically adjust the quantization parameters to realize reasonably the trade-off between coded bits and distortion perception. The experimental results show that the proposed algorithm can save an average of 7.46% and 6.53% bitrate with the same HDR-visible difference predictor-2.2(HDR-VDP-2.2 ) and PSNR~~DE compared with HEVC Main 10, saving the maximum of 18.52 % and 11.49% respectively. The proposed algorithm can effectively reduce the consumption of the overall bitrates and still maintain the visual quality of the reconstructed HDR video.