This PDF file contains the front matter associated with SPIE Proceedings Volume 11915, including the Title Page, Copyright information, and Table of Contents.

Air refractive index measurement technology based on laser interference has shown great application value in many fields, especially in laser ranging, precision manufacturing and integrated circuit manufacturing equipment. With the help of air refractive index measurement technology, the measurement problems in many fields can be solved. In this paper, the research progress of indirect measurement method, synthetic pseudo wavelength method, F-P cavity method, variable cavity length method and optical frequency comb measurement method are introduced, and the advantages and disadvantages of various measurement methods are analyzed. Compared with the indirect measurement method, the direct measurement method based on laser interference technology has higher measurement accuracy. Finally, the current application status of air refractive index measurement technology is introduced, and its future application prospect and development trend are described in detail.

In terms of products, packaging appearance reproduction is the research direction, and in order to effectively guarantee the performance of the reproduction method, this paper uses visual communication technology as the basis and combines laser 3G printing technology to design a packaging appearance reproduction method. The two calibration methods are specifically analyzed, and the corresponding matching points are obtained in the initial image of the package. After effective correction processing, all matching points are minimized while ensuring that the image meets the limit geometric constraints. The result calculates the corresponding Sampson error value and further complete the calculation of the system equations to provide guarantee for the standardization of the final corrected image. Based on this, it gives full play to the advantages of laser 3D printing technology, clarifies the internal structure of the image and the related concepts of the stereo matching algorithm, and obtains the similarity measurement under the comprehensive consideration of various factors, streamlines the scene cloud, and performs the edge part of the image. At the same time, it can control the number of reconstructed point clouds, and finally complete the appearance of the packaging. The simulation results show that for image correction, the time-consuming and operating efficiency are 15s and 100%, respectively, suggesting that the method in this paper has good practicability.

With the continuous development of space technology, the requirements for the technology of modeling for infrared radiation of global background are becoming higher and higher. In this paper, a novel three-dimensional model for the infrared radiation of the global background is proposed, which is rich in elements and detailed information. In the modeling, it first uses latitude and longitude for fine mesh division. Then, this model not only comprehensively consider the terrestrial, oceanic and other elements of the surface of the earth, but also consider aerosols, molecules, the clouds with various types and other elements which are in the atmosphere. In addition, the characteristics of infrared emitted and reflected radiation of these elements under different area, elevation, time and season were researched. Finally, infrared images of global background were simulated, and the results was quantitative analyzed to discuss the model precision. The experimental results show that the proposed model can truly reflect the infrared radiation characteristics of global background, which could provide basic research reference for the studies on space target detection and remote sensing data analysis, etc.

Six degree of freedom laser measurement technology has many advantages, such as high measurement accuracy, wide measurement area, non-contact measurement and so on. It is a powerful multi degree of freedom simultaneous measurement technology, and has very important applications in many fields, such as aerospace, precision manufacturing, weapons and equipment. This paper briefly introduces the main technology of six degree of freedom laser interferometry system, and summarizes the main measurement technology, including the measurement technology and method based on the combination of laser interference and laser collimation, the measurement technology and method based on laser grating interference, and the measurement technology and method based on laser tracking. This paper summarizes the research progress and development of six degree of freedom laser measurement technology in recent years, summarizes the role of six degree of freedom laser measurement technology in the main fields, and forecasts the future development trend of six degree of freedom laser interferometry technology.

Two spatial coupled different lasers with different parameters are studied and their physical model is presented while the two coupled lasers are very interested to show a great nonlinear dynamical variety using bifurcation diagrams of the two lasers. We study dynamics and focus on bifurcation of the two coupled lasers. We show how the bifurcation of the two lasers to vary as the main parameter shift while a route to chaos from bifurcation can be analyzed by varying one parameter of the two lasers, and we analyze what kind of effects of the coupling factor, the current, the photon loss rate, the linewidth enhancement factor and the frequency detuning on dynamics behaviors of the two lasers. Presented bifurcation diagrams show all processes of bifurcation, passing stabilization to period-doubling, unstabilization, chaos, quasi-period. We discuss how the two lasers becoming to go away from chaos to evolve to a stably-state via varying one parameter of the two lasers, such as the lasers becomes of going away from chaos to follow a route to a period-doubling after a quasi-period and a period-tripling. And we point out bifurcation point and find a stable region, a periodic region, a double-periodic region, quasi-periodic region, an unstable region, a chaotic region. The obtained result is of great significance to study laser dynamics, nonlinear optics, chaos of coupled lasers, and ultra-wideband signal generator.

Orbital angular momentum (OAM) of vortex beams can extend the capacity and spectral efficiency of free-space optical (FSO) communication for its orthogonality provide an extra dimension. However, atmospheric turbulence (AT) will lead to wave-front distortion, crosstalk between OAM modes and eventually reduce the communication system performance. This paper uses a Phase Diversity (PD)-based adaptive optics (AO) schema to compensate the OAM beam. With the General Regression Neural Network (GRNN), the relationship between the intensity distribution and the wave-front distortion of propagation in AT channel is established, and the wave-front aberration is retrieved. This method is simple in structure without phase sensor and probe beam, and reduces computation compared with the common phase retrieval algorithms. Therefore, it can compensate OAM beam in a second and ensure the similar retrieve effect.

Photon counting imaging technology has been widely used in nuclear radiation detection, space environment detection, astronomy observation, nuclear physics and ultra-weak bioluminescence. In this paper, the progress and parameters of NUV (160-320nm) and FUV (102-170nm) photon-counting imaging detectors were described. The NUV detector was developed based on a sealed MCP-image intensifier which comprises input window, photocathode, MCP stack, Ge-layer and its ceramic substrate. To maximize the quantum efficiency, a Cesium Telluride (Cs₂Te) photocathode was adopted, which was deposited on input window and mounted close to the MCP. For the FUV detector, because of the lower cut-off wavelength, there are no suitable window materials in this band and the open-faced design should be used to meet the requirements of the detection. Therefore, a Cesium Iodide (CsI) photocathode deposited on the input surface of the MCP was used to optimize detector efficiency. By using an existing wedge and strip anode (WSA), the imaging performance of the NUV and FUV detectors was tested respectively. Experimental results show that the quantum efficiency of Cs₂Te is 12.1% (at 230nm), the spatial resolution of NUV and FUV detectors is better than 70μm, the dark count rate of NUV and FUV detectors is about 10.5- and 2.3-counts/s*cm² respectively.

The synchronization of complex dynamic systems is a hot topic in scientific research. Thus, this paper focuses on the synchronization of two different laser systems and studies the multiple-dynamics behavior of two coupling lasers synchronizing with that of another laser, and present a physical model of the multiple-dynamics synchronization. Two routes to chaos after bifurcation from two stable states are illustrated by controlling one current of the two coupling lasers while the two coupling lasers are found to produce great nonlinear dynamics behaviors via their bifurcation diagrams. And this paper discusses the detailedly multiple-dynamics distribution of the two coupling lasers, where there is a stable-state region, a single-period region, a quasi-period region, and a chaotic region. Our work performs how multiple-dynamics of the two coupling lasers synchronize with that of another laser by controlling one current of the two coupling lasers. Then, multiple dynamics behaviors of the two coupling lasers synchronizations are obtained with that of another laser, such as a single-period synchronization, a period-doubling synchronization, a period-4 synchronization, a period-5 synchronization, other quasi-period synchronization, and a chaotic synchronization. The obtained results are of great significance to study laser physics, optics, synchronizations of two optical dynamic systems, a chaotic laser, and its application.

In view of the existing pedestrian detection methods based on deep learning with low resolution and small pedestrian size, such as false detection, missed detection and low detection accuracy, a small-scale pedestrian detection method based on feature fusion refinement and improved Faster R-CNN was proposed. A fusion and refinement strategy that enhances the ability of multi-layer feature expression is used to improve FPN, and a variable-shape anchor is constructed in RPN to extract pedestrian candidate regions, in order to reduce the influence of background on detection results. In addition, a feature adaptive module is added to the detection network, and the shape information is incorporated into the feature map to adapt to the anchor with variable shape. The experimental results show that compared with Faster R-CNN, the proposed method has an improved effect on the data sets INRIA and CityPersons under the condition of equivalent detection efficiency.

Infrared polarization image has excellent applications in target detection, tracking and warning, etc., but the original infrared polarization image has problems such as cold reflection and line blurring, which results in the low quality of the final image and affects the overall effect of infrared polarization detection. To address the above shortcomings, we propose an improved image correction method based on single-pixel non-uniformity differential image correction to remove the cold reflection. In addition, we use the two-dimensional discrete cosine transform method to Sharpen the image. The results show that the proposed method can effectively suppress the cold reflection problem and improve the robustness of the image.

With the popularity of high-precision printing equipment, leading to the widespread spread of pirated products. Because of some disadvantages in traditional anti-counterfeiting printing, a half-tone anti-counterfeiting image generation algorithm based on microstructure dots is proposed. First, the word library automatically generates the character word library containing the grayscale and morphological differences, and then matches and replaces the image according to the similarity of the image, so as to realize the generation of the anti-counterfeiting image. Experimental results show that the anti-counterfeiting image based on microstructure dot generated by this algorithm is visually small from the original image, and after the anti-counterfeiting image is enlarged, the embedded micro characters are clearly visible, and the algorithm has broad application prospects in packaging anti-counterfeiting.

Visual extraction of railway is widely used in automatic driving and obstacle detection. At present, there are many research results on lane line detection. However, there is a big difference between lane line detection and railway extraction. Rail extraction faces many challenges, such as complex environments and complex rail shapes. This paper introduces a railway extraction method based on sequential line detection. In this method, picture is divided into multiple stages from bottom to top. According to proposed matching score, select lines that are most matching to railway in each stage are selected. Then, all detected lines are linked as final extracted railway, which makes proposed method is effective to straight railway and curved railway. Experiments have proved that proposed method remains accurate and stable in different environments.

The dual-camera coded aperture snapshot spectral imager(DC-CASSI) includes the coded aperture snapshot spectral imager(CASSI) and panchromatic imager. CASSI can obtain the three-dimensional spectral image information of the target in a single coding within a single integral time, but it is difficult to achieve high-quality reconstruction of spectral image in a single coding. Therefore, the panchromatic image acquired by the panchromatic imager should be fused with it to obtain high-quality multispectral reconstruction images. Based on the imaging characteristics of DC-CASSI, a multispectral and panchromatic image fusion algorithm based on Non-subsampled Shearlets Transform (NSST) and improved Pulse Coupled Neural Network(PCNN) is proposed. The fusion experimental results show that compared with other traditional fusion algorithms, the proposed fusion algorithm can be well applied to DC-CASSI and maximum improving the spatial resolution of multispectral coded image while preserving spectral characteristics of the multispectral coded image.

Millimeter wave and terahertz imaging technology is a frontier detection technology with practical application values for people's livelihood, national defense and other fields. This paper summarizes millimeter wave and terahertz imaging technology, consisting of its technical principle, historical development, research status and practical applications.In the paper,discusses focus on the applications of millimeter wave and terahertz imaging technology in security inspections.

Precision guided weapons are the main means of physical destruction in information warfare, and are the key to winning future wars. Infrared target recognition is an important factor affecting its accuracy. Because the imaging mechanism of infrared image and visible light image is different, there is a big difference between the two in the gray level and edge information of the corresponding area. This paper proposes a fusion of gray matching and feature matching algorithms to enhance the reliability of the results. Experiments have proved that by fusing the results of grayscale matching and feature matching, the grayscale information and feature characteristics of the image can be fully utilized to correct the mismatch phenomenon in infrared image target recognition. Compared with traditional single gray level matching or feature matching, it has more accurate matching results, and has certain application value for typical target matching and recognition.

Accurate acquisition of the canopy information of fruit trees is very important for the precise variable spray of modern orchards. Using drones to obtain fruit tree canopy images is a better nondestructive method, but the lighting conditions in the orchard are complex. It is difficult to quickly extract the canopy of fruit trees from the aerial images of orchard by UAV, which can be used to guide UAV to apply pesticide in real time and accurately. Therefore, we propose a method for canopy segmentation of fruit tree canopy images using SegNet network model. We use images obtained from modern orchards to verify the accuracy and real time of the network, and use four indicators to compare our network model with Unet and FCN-8s network models: accuracy, precision, recall and harmonic average. After that, we optimize the SegNet network model structure from three aspects: input method, network training parameters and neural network structure. The results show that SegNet has achieved satisfactory results in segmenting the canopy. The optimized SegNet model has an average recognition accuracy of 95.30%, and the recognition time of a single image is as low as 0.045 s, and it has good robustness in both strong and weak light environments. This shows that using SegNet network segmentation to extract fruit tree canopy information is a promising method, and it can provide a reference for real-time and accurate spray of UAVs.

In the wafer manufacturing process, more and more mixed-type wafer maps have appeared. This paper proposes a classification structure of mixed defect types wafer maps based on deep convolutional neural networks. First, synthesize wafer maps of mixed defect types in actual production through simulation, and then use the deep convolutional neural network model to construct a classifier based on the wafer map of the basic defect types, and train the classifier. Finally, the classification results of all basic classifiers are combined to determine the actual mixed defect type of the wafer map. The final classification results verify the effectiveness of the model.

The forming of the ink droplet is very important in the inkjet control system. It is related to the forming of the ink droplet on the substrate with good printing effect. In the process can use of inkjet ink observation system to observe the ink drops of spray forming process, at the same time, use the LED flash and CCD camera to film the formation of the ink droplets, through the pictures use edge detection algorithm to analyze drops forming condition and flight path, still can use the shape of the Hough transform to detect the drops (mainly used to detect circular drops), and finally by MATLAB/Simulink simulation tool to drops control system simulation analysis, The data of ink droplet forming is analyzed and tested in the ink feeding system.

In order to further improve the robustness of the ambiguity suppression related filtering algorithm (ARCF) tracking in many complex scenes such as rapid motion and occlusion, an adaptive normal punishment coefficient is proposed based on the relevant filter algorithm of automatic space-time normalization. At the same time, in order to make the tracking results more accurate, a object peak detection and selection algorithm is proposed specifically for multi-peak situation, which can determine the peak condition of the response map of the current frame and select the peak that best meets the object in multi-peak situation. In the template update stage, in order to keep the updated template from the pollution of similarities and masks, this paper also proposes a situation to detect whether the object is obscured, as a condition for the template update, so that the object template can better match the object.

Image inpainting is a means of inferring the content of the missing information based on the remaining information of the image, and inpainting the missing information, requiring the inpainted image to meet the human eye perception to the greatest extent. The image inpainting method based on the weight function takes arbitrary points for inpainting during inpainting, which results in poor inpainting effect at the junction of objects. Therefore, this paper adds the priority values calculation of inpainting points to improve the inpainting effect. Finally, comparing the inpainted image based on the priority calculation with the previous one, the inpainting effect at the boundary of the object is significantly better than the image inpainting method based solely on the weight functions.

X-ray nanotomography has become an important analysis tool in a wide range of fields. However, the imaging quality is often affected by drift from focal spot movement and mechanical instability. An improved horizontal drift correction method for X-ray nanotomography based on trajectory of sinogram centroid (TSC) is proposed. This method requires neither auxiliary marks nor additional projections. A sliding window TSC fitting method is utilized. The sum of the squared errors (SSE) is calculated between the trajectory and standard sinusoidal curve. The one corresponding to the minimum SSE is chosen to obtain the horizontal drift from the original TSC, which is then used to align the projections. The proposed method is evaluated by both simulation results of the Shepp-Logan phantom and nanotomographic results of the honeybee mouthpart. The results show that this method can quickly and effectively correct the projection horizontal drift.

At present, the welding seam extraction algorithm is for 2D space image processing, which has the disadvantages of complex extraction process, low accuracy and great limitations. Compared with the cumulative error caused by image mosaic, 3D can reflect the target features from various perspectives. At present, RANSAC fitting plane and Euclidean clustering extraction have been applied to 3D point cloud feature analysis, but RANSAC can only roughly fit the surface of the object, and it can not effectively extract features for the areas with rich surface curvature changes. Euclidean clustering is based on KD tree query where the algorithm needs to set the number of neighborhood points, so it has uncertainty. In order to overcome the above defects, this paper proposes a region growing segmentation algorithm, which can divide the region and extract the boundary according to the changes of the weld surface curvature and normal vector. At the same time, compared with the actual extraction results, the average error is reduced to about 1.0 mm. Compared with the traditional 2D laser line extraction, it can more intuitively reflect the weld features and put into practical production application.

Because of its low cost and convenient use, UAV is often used to carry out reconnaissance and detection missions. At present, in the process of UAV carrying binocular camera to identify and locate the target, it often faces the problems of insufficient endurance and easy to be found. On the basis of binocular visual positioning, this paper puts forward the method of cooperative positioning of double UAV. Using a single camera to coordinate positioning when two UAVs fly side by side can ensure that it can still complete the task when a UAV is out of control or falls. In the process of target positioning, the binocular vision positioning principle is used to solve the target coordinates by least square method, and the error analysis is carried out by Monte Carlo method. The simulation results show that the error can be controlled within the acceptable range, and the load capacity and endurance ability of UAV can be increased without affecting the positioning accuracy.

This paper aims to solve the problem of long acquisition time in beacon-less acquisition system for satellite laser communication. This paper splits and expands the Gaussian laser beam, and proposes three laser beam scanning schemes, including simultaneous scanning in areas, simultaneous scanning in the inner and outer circles, and simultaneous scanning in jumps. When one laser beam is divided into multiple laser beams, increasing the beam waist radius of the laser means increasing the beam divergence angle. In order to increase the total effective spot area, multiple laser beams are expanded to scan at the same time without affecting the scanning uncertainty region and the response intensity of photodetector. Accordingly, the number of scanning steps required for acquisition is reduced, thereby shortening the acquisition time and speeding up the successful acquisition. Results show that the proposed beacon-free fast acquisition system based on laser beam splitting and beam expanding can improve the acquisition efficiency of satellite laser communication without affecting the optical detection responsivity.

Nowadays, Quick Response (QR) code technology is used in all aspects of our lives due to the widespread use of smartphones. However, when the Light Emitting Diode (LED) display is suspended from the ceiling, the actual camera is not facing the LED and must have a certain pitch angle when shooting, so the captured image must have a three-dimensional spatial rotation effect, which leads to low decoding rate. Therefore, to address this phenomenon, this paper proposes an image processing method based on the combination of Hough transform and affine transform, which solves the problems of image rotation in the recognition of QR code images at the receiving end.

The ocean covers more than 70% of the global area and contains rich resources. Marine exploration has attracted more and more attention due to its scientific, strategic and economic importance, and one of the necessary conditions for marine exploration and development is the ability to have underwater communication. Underwater acoustic communication is a relatively mature means of communication at present, but due to the influence of multipath interference and Doppler effect, the underwater communication transmission bandwidth is small, the time delay is high, and can not meet the needs of high-speed transmission. Underwater wireless optical communication (UWOC) can provide large bandwidth transmission, low latency and high security. However, the scattering characteristics of seawater to light cause time-domain dispersion, which limits the channel bandwidth. In addition, the influence of optoelectronic modulation and devices limits the bandwidth of receiver and transmitter. In recent years, underwater wireless communication technology has attracted much attention, and its application in the field of ocean engineering has broad prospects, such as underwater high-definition video transmission and deep sea observation, so the realization of high-speed underwater wireless communication system has become an urgent problem to be solved.

The production date on the food package is an important safety information. It is widely used in the production line to print the production date with inkjet code printer. However, character defects such as fracture and displacement may occur due to machine failure and offset of packaging materials. In order to improve the printing quality of packaging characters of production line, this paper studies the related classification algorithm for the common printing defects of production date. In order to classify defects, a 17-dimensional feature vector is designed to describe defects based on the geometric features of defects. Because of redundancy and interference information, linear discriminant analysis (LDA) is used to reduce the dimension of the eigenvectors. The classification mode and parameters of SVM are experimented on the classification section. According to the experimental data, the classification accuracy is more than 94%.