This PDF file contains the front matter associated with SPIE Proceedings Volume 8917, including the Title Page, Copyright information, Table of Contents, Introduction, and Conference Committee listing.

In this paper, we present a new LC lens with multi-ring patterned electrode, it consists of two ring-shaped sub-electrodes and a circle sub-electrode, each sub-electrode can be driven separately. The two rings have different diameter but the same center, when the two ring-shaped sub-electrodes are driven independently, the LC lens can work as micro convex lens with different clear aperture and its local length can be tunable along optical axis by electrical signal. As the voltage is applied to the circle sub-electrode, it appears like concave lens. With this design, it can achieve two types microlens effect in one structure, and by applying the voltage to the different patterned electrode, it can switch between two types microlens. The optical properties of the LC microlens are also demonstrated experimentally.

ISAR (inverse synthetic aperture radar) can generate 2D image of a non-cooperative moving target and be used for military and civilian purpose. For moving target，well-focused ISAR images can be achieved using appropriate motion compensation and image reconstruction algorithm. The ISAR image will be difficultly when the echo is at a low signal-to-noise level. For polarimetric ISAR system which transmits and receives different polarisation signal in both two channels, the fusion of HRRP (high range resolution profile) from different channels will improve significantly the SNR of the signal. The method proposed in this paper firstly fuses the different channels HRRP to get a higher SNR signal. Then the target region is extracted in fused HRRP. Finally, the motion compensation for ISAR imaging is carried out only on the extracted target region data in the sense of using image reconstruction method to gain a focused ISAR image. The real measured data shows the validity of the algorithm.

This paper proposes a method to characterize a liquid crystal Fabry-Perot (LC-FP) hyperspectral imaging devices with electrically controlling and shifting wavelength of light beam penetrating the LC-FP device proposed. As the effect of the directing vector distribution of LC materials can be remarkably changed and stably anchored in an electric field applied over the electrodes of the device, the refractive index of the LC materials which filled into FP cavity can be modulated by voltage signal with relatively low amplitude. We achieved the transmissivity properties of the infrared radiation out from the LC-FP and continuously incident upon the photosensitive structure through the thin film matrix equation, and then acquired the basic relationship among the transmissivity, and the applied voltage signal, and the wavelength parameter selected. The key features of the proposal approach is that the device is mainly composed of very thin electrically modulating refractive-index architecture, which are generally two basic FP interferometers connected closely so as to accurately perform the choice of the wavelength desired in the spectral range by the interference filtering; and the depths of the two FP interferometers are designed to be different, and through combination two parts mentioned above we can obtain some single wavelengths desired of penetrating beam. The proposed device has already been complished. The final device will be used to sequentially choose each desired infrared (IR) wavelength from incident IR radiation. Prototypes of the LC-FP with 4×4 arrays is going to be made, thus we can get several different wavelengths with the device.

Bad pixels and response non-uniformity are the primary obstacles when IRFPA is used in different thermal imaging systems. The bad pixels of IRFPA include fixed bad pixels and random bad pixels. The former is caused by material or manufacture defect and their positions are always fixed, the latter is caused by temperature drift and their positions are always changing. Traditional radiometric calibration-based bad pixel detection and compensation algorithm is only valid to the fixed bad pixels. Scene-based bad pixel correction algorithm is the effective way to eliminate these two kinds of bad pixels. Currently, the most used scene-based bad pixel correction algorithm is based on adaptive median filter (AMF). In this algorithm, bad pixels are regarded as image noise and then be replaced by filtered value. However, missed correction and false correction often happens when AMF is used to handle complex infrared scenes. To solve this problem, a new adaptive bad pixel correction algorithm based on pulse coupled neural networks (PCNN) is proposed. Potential bad pixels are detected by PCNN in the first step, then image sequences are used periodically to confirm the real bad pixels and exclude the false one, finally bad pixels are replaced by the filtered result. With the real infrared images obtained from a camera, the experiment results show the effectiveness of the proposed algorithm.

The liquid crystal (LC) device with the circle aperture electrode can be used as a convex lens. The index of refraction of the LC lens can be easily changed by the voltage signal; the arrayed LC lens can be used in the imaging sensors. Comparing with the traditional optical lens, the LC lens has a more effective architecture. In this paper, we present an imaging sensor with a new type LC structure composed of three layers of top electrodes and a joint bottom electrode, and simulating the tunable spatial resolution architecture carried out by applying voltage signal over different ITO electrode in LC lenses coupled with arrayed imaging sensors. From the result of the simulation, we can find that changing the spatial resolution by our architecture can be achieved.

For quick detection of “open field” targets and landmines in tussock grass over wide area AMUSAR system must operate in low frequency below 500MHz and high frequency above 2.5GHz. The previous system calibration method used in AMUSAR adopts a compromise in single calibration object such as 0.1m trihedral. But for the frequency band between 500MHz and 1GHz, the 0.1m trihedral locates in Relay region with low signal-to-noise ratio (SNR). In order to improve performance of penetration, we make some modifications as follows. The data between 1GHz and 2.5GHz is calibrated by 0.1m trihedral. Since the 0.3m trihedral exhibits small fluctuation and large RCS between 500MHz and 1GHz, it ensures the SNR of calibration data. So the data is calibrated by 0.3m trihedral in this frequency band. The results show that the proposed method calibrates the system errors more efficiently and thus enhance the performance of imaging.

In this paper, we design and implementation anti-reflection and high reflection film for tunable LC-FP filter working in typical medium wave infrared (MWIR) spectral region. As to reflection mirror, we complete it by depositing gold film on the silicon substrate. By using the analytical method of dividing amplitude multiple beam interference to simulate the reflectance, the reflectivity result of the gold film is around 98% at the average in the MWIR spectral region. When take the absorption of the gold film into consideration, the gold film should be thin under the condition that it is conductive well. The anti-reflection film is introduced to reduce the reflection of the other side of the wafer. In anti-reflection structure, we simulate the reflection of the films with the algorithm of the equivalent membrane and fulfill our design with the technology of vapor deposition. Finally, we test the total transmittance of the wafer deposited gold films and anti-reflection films, which is about 0.2% of single chip. By making use of the wafer designed by us, we fabricate the LC-FP filter by placing two wafers side by side with the anti-reflection structure face to the direction of wavelength incident. Simultaneously, the LC layer with fixed thickness is sealed between the two high reflection mirrors formed by gold film. Compared with other method to fabricate mid-infrared FP filter, deposition of reflection and anti-reflection films on wafer have the advantage of low cost, simple technology.

In order to form a false scene in Synthetic Aperture Radar (SAR) image, deceptive jammer need to get the relevant SAR parameters. In these parameters, squint angle and beamwidth usally change and it will make the pre-generated jamming signal unuseful. For solving this problem, a strategy is proposed to transform the pre-generated jamming signals to counter SAR with arbitrary squint angle and beamwidth in real time. Firstly, the jamming effects under estimation errors of SAR’s squint angle and beam-width are analyzed. Using Graphics Processing Units (GPU), a parallel algorithm to generate jamming signals for varying squint angle and azimuth beam-width is proposed. Then, This paper describes a method that can implement the signal transformation between wide-beam condition and narrow-beam condition. Based on the generated signals, the jamming under arbitrary squint angle and beam-width can be realized in real time. The simulation results shows that this strategy is effective to jam SAR with varieties of squint angles and wide-beams.

In this paper, an integrated hyperspectral imaging sensor array technology by using arrayed diffractive micro-optics elements for driving infrared beams with chosen wavelength into designated subsensors is proposed. The diffractive optical element, which can be treated as a functioned microlens here, collect the incident Gaussian beam or other types of infrared beams out-from targets and then concentrate the incident light into desired monochromatic point pattern or arbitrary distinct hyperspectral image at the imaging plane based on phase transformation and diffraction propagation process. For the incident infrared beams with different wavelengths, the arrayed diffractive micro-optics elements is designed based on the diffractive integral theory, and the weighted iterative phase retrieval algorithm is modeled so as to obtain the needed phase distribution, and therefore the frequency spectrum of the incident beams can be separated locally in different locations over the focal plane according to designated wavelength. We can then achieve beam-splitting imaging by placing sensor array (4×4 pixels per subsensor) on the focal plane at locations that correspond to different wavelengths. Simulation results demonstrate that the designed elements can successfully implement both the functions of demultiplexing different wavelength beams and focusing each component at a pre-designated position simultaneously.

High speed movement of target leads to quadratic phase of the dechirped signal. A new high speed movement compensation method based on integrated cubic phase function is proposed. Numerical result shows that the algorithm can precisely estimate the quadratic phase coefficient and compensate the high resolution range profile under low signal to noise ratio. After compensation, the aberration of range profile is revised and the quality of ISAR image is greatly improved. In this way, the effectiveness of the method is verified.

In this paper, a three-dimensional relaxation method and a finite difference method (FDM) are used to model the dynamic response behavior of liquid crystal (LC) microlens. Simulations have been done for the focus-swing patterned electrodes structures. The formation of disclination lines in the director orientation can be accurately predicted. Based on the simulation, a careful choice of the device structure and the voltage will help to design better lens. We can expect that the model developed by us can be utilized to design more complex LC microlens or other functioned LC structure.

Imaging detector research on terahertz (1 THz=10¹²Hz) wavelength region is a hot topic in recent years. So far, the terahertz wave has shown considerable application potential in advanced imaging of some special targets¹. The terahertz imaging has a relatively high resolution correspondent to common radio frequency imaging. It takes object surface shape and internal structure information², which can not be captured by current mature optics imaging means or conventional infrared imaging methods. However, for a long time, due to the lack of effective terahertz detection techniques³ as well as most materials in nature inherently do not respond to THz radiation, or demonstrate a very weak response even not be utilized, it limits the development of terahertz theory and related technologies and applications especial in the terahertz detection fields, and thus leading to the progress and applications in THz regime lagging behind the rapid development of other electromagnetic spectrum. To realize the useful potential detecting applications of THz radiation, considerable efforts are underway for filling the ‘THz gap’. Some new type of materials such as typical metamaterial are really needed for constructing detecting architectures.

Metamaterials⁴, which are artificially structured material, consists of subwavelength metallic resonators within or onto a dielectric or semiconducting substrate. Research and applications^5,6 show that they already exhibit attractive electromagnetic properties, which are not available in naturally materials. Therefore, they can be used to enhance the optoelectronic response ability so as to efficiently manipulate, control, and detect electromagnetic radiation by particularly designed metamaterial micro-nano-structure. In addition, by scaling their size, we can scale their response from radio frequency to optical wavelength region, which means we can design metamaterials detector operated at desired frequency in a very wide frequency range, for example from UV-IR-Microwave-RF, but in THz region is our key topic content in this paper.

As demonstrated that metamaterials can be used to remarkably response incident THz radiation with both electric and magnetic resonant mode⁷, metamaterials micro-nano-structures are potential for future THz detection. By constructing metamaterials detectors for relatively wide THz wave, we can realize THz sensing and further imaging. In this paper, we design and simulate an electrically resonant terahertz metamaterial sensing unit. This kind of metamaterial micro-nano-structure can present an obvious response at 0.78 THz with a strong electrical resonant at the split-ring resonator（SRR） gap, and thus provides a possibility to obtain the electrical signal so as to achieve THz sensing. By analyzing the simulation results, we summarize the feasibility of terahertz detection, and come out a layout of terahertz detector by scaling the size of metamaterial detector unit, we can obtain unit detector architecture that also resonant at other frequencies and finally lead to realize multispectral imaging.

This paper introduces a coaxial visible and infrared dual-band imager, which utilizes the visible and infrared detection technology, uptakes the scenery radiation of different wavelengths or optical energy reflected, adopts transmission and secondary reflection theory to design the coaxial optical path, and realizes the dual-band imaging of same scene. The imager can acquire the registered visible and infrared images, and effectively solves the problem of registration for visible and infrared images in different-source image fusion.

Target recognition in natural scenes generally uses the remote sensing image preparation for the matching feature template. Image match system finds correspondence between real image and remote sensing image and then output the target position parameters to the aircraft. How to quantitative analysis and evaluate the robustness of the infrared features, and to determine the available features in infrared image matching recognition algorithm is one of the keys of the ground target recognition in complex scenes. In this paper, we built a feature robustness evaluation model for typical match identifying by analyzing for the robustness of features extracted from typical terrains and targets in various condition. Combined with France
SE-Workbench-IR simulation platform, we designed a special scene simulation development process, in case of lack of terrain generation module, it also can generate MWIR natural scene image. By analyzing the simulation image and real-time image in the same condition, we can gain the variation information from infrared radiation characteristic in different natural condition. Finally, we verify and assess the robustness of the matching features.

Optical retification of laser pulses in LiNbO₃ by tilted-pulse-front pumping(TPFP) is a powerful way to generate terahertz(THz) pulses. Detailed theoretical analysis of the TPFP THz pulses generation efficiency and polarization with different pump beam polarization was presented. The THz generation efficiency for various of pump beam polarization angle and THz pulse polarization were also researched. The results predict that the magnitude of the THz electric field is maximized when the pump beam electric field vector is parallel to the LiNbO₃ crystal axis.

The paper proposes a modularized infrared scene imaging simulation solution based on GPU and designs a corresponding implementation framework to contain all the key modules of infrared imaging simulation. After that, the paper presents the module of infrared radiation computation, which realizes the real-time parallel computation of infrared radiation by the technology of GPU shader and translates the simulation work from 3D graphics to 2D digital image by the technology of rendering to texture.

Digital cameras are now more than 30 years old. They benefi from an exceptional popularity that the conventional _lm cameras never obtained. Digital cameras differ from conventional analog cameras with respect to several distinctive features, some of which stand as undeniable improvements, some remaining on the contrary as brakes towards their use in every technical field. However, in the last years, these distinctive features are fostering new techniques of world observation, these techniques relying merely on the openings proposed by the new sensors. Here is the domain of Computational Photography. Before we present a critical survey of the objectives of computational photography, a review of the distinctive features of digital cameras is made, in order to determine where progress may be expected.

It is very difficult to detect a moving object then estimate its speed with optical remote sensing images. In this paper, we propose a novel method that moving object detection and velocity measurement using only single multispectral image. Slight chromatic aberration among channels of the multispectral image is applied to detect the motional object by fames difference, and improve the accuracy of velocity measurement using pixel unmixing. Finally, experiments are carried out on satellite multispectral images from Google Earth and clearly show the comparative effectiveness of the proposed method with respect to traditional methods.

Image and video quality measurements are crucial for many applications, such as acquisition, compression, transmission, enhancement, and reproduction. Nowadays, no-reference (NR) image quality assessment (IQA) methods have been drawn extensive attention because it does not need any information of reference images. However, most proposed NR IQA methods are designed only for one or a set of predefined specific distortion types, which are unlikely to generalize for evaluating images distorted with other types of distortions. In order to estimate a wide range of image distortions, in this paper, a novel NR IQA method is proposed which is based on shearlet transform, a new multiscale directional transform with a strong ability to localize distributed discontinuities. The distorted image leads to significant variation in the distributed discontinuities in all directions. Thus, the statistical property of the distorted image is significantly different from that of natural images in shearlet domain. A new model is also proposed to measure this difference. Numerical experiments demonstrate that this new NR IQA method is consistent with subjective assessment, very effective for many well-known types of image distortions and superior to some existing prominent methods.

Ship detection is significant especially with the increasing worldwide cooperation in commerce and military affairs. Space-borne Synthetic Aperture Radar (SAR) is optimal for ship detection due to its high resolution over wide swaths and all-weather working capability. Constant False Alarm Rate (CFAR) detection of ships in SAR imagery is a robust and popular choice. K distribution has been widely accepted for homogeneous sea clutter modeling. Although localized K-distribution based CFAR detection has been developed to solve the non-homogeneous problem, it is not satisfied under adverse conditions, for example, interference target appears in the background window. In order to overcome its shortcomings, this paper presents an adaptive algorithm to improve the performance. It mainly includes the homogeneity assessment of the local background area and the automatic selection between the localized K-distribution-based CFAR detector and the OS-CFAR detector, which has better detecting performance in non-homogeneous situation. The theory is investigated in detail firstly, and then experiments are carried out and the results illustrate that the novel algorithm outperforms the state-of-art methods especially under complex sea background condition.

A watermarking algorithm of color images based on the quaternion Fourier Transform (QFFT) and improved quantization index algorithm (QIM) is proposed in this paper. The original image is transformed by QFFT, the watermark image is processed by compression and quantization coding, and then the processed watermark image is embedded into the components of the transformed original image. It achieves embedding and blind extraction of the watermark image. The experimental results show that the watermarking algorithm based on the improved QIM algorithm with distortion compensation achieves a good tradeoff between invisibility and robustness, and better robustness for the attacks of Gaussian noises, salt and pepper noises, JPEG compression, cropping, filtering and image enhancement than the traditional QIM algorithm.

A procedure for haze detection and removal from a single image using dark channel and fast bilateral filter has been developed. It involves the analysis of the statistics of the haze images. The dark channel prior with the haze imaging model is used to estimate the parameters of the haze and recover a high quality haze-free image. With the fast bilateral filter, we can acquire the fine transmission estimation and has low computation. Experimental results on variety of hazy images demonstrate the proposed method is simple and powerful.

The anomaly of surface object associated with underground nuclear explosion (UNE) was regarded as detected target using remote sensing image, because of its unknown geometrical and spectral feature, it is difficult to detect anomaly using general methods of change detection based on gray and texture feature using remote sensing data. A multi-feature joint matching method of anomaly detection was developed based on spectral matching. Firstly, taking pixel spectral curve of image acquired before the UNE as reference spectral vector, and the one acquired after the UNE as matched spectral vector, then the similarities of spectral shape, feature and value were calculated through spectral matching algorithm. Secondly, each similarity above was normalized and weighted by appropriate coefficient; the multi-feature joint matching result was produced by accumulating the weighted similarity. Finally, appropriate segment method was chosen to distinguish target from background, and the pixels with high unmatched degree (namely anomaly) were detected. We took some UNE events for example, the anomaly of the UNE was detected effectively, and the problem of undetected anomaly just using single spectral matching algorithm is well resolved, the result also suggests the method be in good generality.

Resistor array plays a vital role in emulation of the IR control and guide system. However, its serious nonuniformity confines the range of its application. Therefore, in order to obtain an available IR image, nonuniformity correction (NUC) is necessary. The traditional method is sparse grid and flood which only take the array’s nonuniformity into account. In this paper we present an improved sparse grid method which considers the whole system which affects the array’s nonuniformity by dividing the NUC process into different gray levels. In each gray level, we can take two points or several points to calculate the nonuniformity of every block which is divided before correction. After that, we can have several characteristic curves which will be operated with curve fitting. By this means, we will correct the nonuniformity. At last, through the experiment of a number of images, we find the residual nonuniformity associated with random noise is about 0.2% after the correction.

Blind motion deblurring from a single image is a challenging ill-posed problem. Significant progress has been made since blur kernel estimation method using salient edge prediction on transfer region is proposed. However, as selection rule for points to estimate the blur kernel has not been researched deeply, some texture and noise points were taken into account for blur kernel estimating, which makes the existing methods not robust enough. This paper propose a robust motion deblurring algorithm using salient edge prediction on transfer region, which employs a new metric to select transfer region points for kernel estimation. A novel kernel refinement method with hysteresis thresholding is also proposed and adopted by the algorithm to reduce the kernel noise. Extensive experiments show that the algorithm achieves good results, while both the new metric and the novel kernel refinement method improve robustness of the restoration algorithm.

Filtering is a key way to classify ground and non-ground points from LiDAR data. Current LiDAR filtering methods, implemented mostly through single LiDAR point cloud only, lack real data source references, subsequently there may be many errors. In view of this situation, this paper presents a new approach based on topographic change detection for airborne LiDAR filtering. In this method, LiDAR point cloud data are accurately registered with existing DEM data and harmonization to the same coordinate system. Then, we implement grid-based organization of LiDAR point cloud data to make corresponding regions of LiDAR point cloud data and existing DEM data correspond to each other. Finally, we filter LiDAR data based on topographic change detection. In order to verify the algorithm, two types of data with different terrain characteristics-urban and mountain were tested. The experimental results indicate that the approach put forward in this paper is effective in filtering out non-ground points from both urban and mountainous areas, while retaining topographic details.

Spatial smoothing filters are usually used in LMS-based non-uniformity correction methods to get an estimation of true scene. Edge-smearing of these filters is the main reason for ‘ghosting’ artifacts. A method based on sigma filter is proposed, which can improve accuracy of non-uniformity and mitigate ‘ghosting’ artifacts. Sigma filter is also used to detect abnormal pixels and adaptively adjust learning rate. Tests with simulated data and real infrared sequences have shown that the proposed method outperforms other representative LMS-based methods.

A multi-layer coding algorithm is proposed for grey image lossless compression. We transform the original image by a set of bases (e.g., wavelets, DCT, and gradient spaces). Then, the transformed image is split into a sub-image set with a binary tree. The set include two parts: major sub-images and minor sub-images, which are coded separately. Experimental results over a common dataset show that the proposed algorithm performs close to JPEG-LS in terms of bitrate. However, we can get a scalable image quality, which is similar to JPEG2000. A suboptimal compressed image can be obtained when the bitstream is truncated by unexpected factors. Our algorithm is quit suitable for image transmission, on internet or on satellites.

A remote sensing image classification algorithm based on image activity measure is proposed, which is used for adaptive image compression applications. The image activity measure has been studied and the support vector machine(SVM) is introduced. Then, the relationship between the image activity measure and the distortion caused by quantization is discussed in our image compression experiments (JPEG2000, CCSDS and SPIHT). Another two image activity measures are proposed as well. Then a feature vector is constructed by image activity measures in order to describe the image compression features of different images. The test images are classified by support vector machine classifier. The effectiveness of the proposed algorithm has been tested using an image data set, which demonstrates the advantage of the proposed algorithm.

Images are usually degraded by haze and fog due to atmospheric absorption and scattering. Especially for remote sensing, degraded images which suffer from haze will highly probably lose detailed information. Nowadays researches on haze removal mainly focus on visible images. However, as infrared imaging is more and more widely used in many fields, a comprehensive study of the effects of fog on infrared images is urgent. In this paper, we firstly introduce that the obvious attenuation effects of fog on infrared images by analyzing the mid-infrared images captured under different weather conditions. Therefore, we propose a unified dehazing approach for infrared images. The proposed approach mainly includes three steps. The first step is to obtain the local transmission of the original hazed images by using the statistical prior knowledge known as “dark channel” which is similar to visible images. With local transmission, the refined transmission map is estimated in a soft matting framework. Due to atmospheric scattering model, dehazed infrared images can be recovered finally. From experimental results, it is quantitatively demonstrated that the proposed approach can significantly improve the quality of hazed images by using principle component analysis technique.

Natural images often suffer from the problem of noise. In this paper, we present a new denoising method which combines the adaptive weighted median filter with traditional median filter. Inspired by the image segmentation algorithm based on transition region, an image matrix based on the synthesized of local entropy and local variance is calculated. The values of matrix reflect the frequency and intensity of the gray level changes in the neighborhood windows. On the basis of the values of the matrix, the filtering strategy is that the traditional median filter acts in non-transition region, the adaptive weighted median filter acts in transition region, and the weights are set by the values of the matrix too. The major novelty of the proposed algorithm is that it can adequately utilize the advantages of the two filter methods above. Experimental results show that the proposed method outperforms the conventional methods in removing noise effectively and preserving image edges and details thus is suited for natural images denoising.

According to SAR imaging mechanism of underwater bathymetry, a 3-D hydrodynamic model based on SELFE (Semi-implicit Eulerian-Lagrangian Finite Element) is applied to demonstrate the temporal variations of the channel bathymetry changes. A microwave radar imaging of oceanic surface’s program of M4S is used to simulate the variation of normalized radar cross section (NRCS) induced by the ocean surface current. The simulation is carried out to study the effect of water depth changes and current variations on SAR imaging of channel bathymetry. The depth of channel is defined between 5 m to 17 m with an interval of every two meters. Six time of tide are used in the current variation simulation. The NRCS variation demonstrated in the simulated images are compared with in situ data, historical bathymetric maps and calibrated SAR images. All the four kinds of images manifest similar changes of channels, which proves that there is agreement between the simulation model used in this paper and other data. Simulation results also show that bigger depth change induces bigger NRCS variation. In the time of maximum ebb tide, simulated relative NRCS is bigger than others.

The coordinate systems related to the vehicle-based mobile laser scanning system and the relationships between them are firstly demonstrated in this paper, then the positioning principals of vehicle-based mobile laser scanning system are well established. Factors that influence the system positioning precision are analyzed briefly, including the misalignment angle and lever arm between Scanner Owned Coordinate System (SOCS) and IMU System. To calibrate the misalignment angle and lever arm between SOCS and IMU Coordinate System, a calibration field, which uses the building corners as reference points, is established. Based on the calibration field established, a calibration method is proposed and utilized in the calibration of our vehicle-based mobile laser scanning and urban panoramic imaging and surveying system, the verification experiment show that, under good GPS conditions and with the system well calibrated, the elevation accuracy is better than 4cm and the planimetric accuracy is better than 6cm.

It is rather difficult for low-level visual features to describe the need for specific applications of image understanding, which results in the inconsistency between vision information and application need. In this paper, a new model is proposed to reduce this gap by combining low-level visual features with semantic features. It uses the output of neural network as the semantic feature, which is accompanied with the priori label features to describe the image after making normalization. And then, the proposed method employs Potts to model the distribution of label priori, and utilizes the Bayesian network to classify images. Several experiments on both synthetic and real images have verified that this method can get more accurate segmentation.

The clustered DPCM (C-DPCM) lossless compression method by Jarno et al. for hyperspectral images achieved a good compression effect. It can be divided into three components: clustering, prediction, and coding. In the prediction part, it solves a multiple linear regression model for each of the clusters in every band. Without considering the effect of noise spectra, there is still room for improvement. This paper proposes a C-DPCM method with Removing Noise Spectra (C-DPCM-RNS) for the lossless compression of hyperspectral images.
C-DPCM-RNS's prediction part consists of two-times trainings. The prediction coefficients obtained from the first training will be used in the linear predictor to compute all the predicted values and then the difference between original and predicted values in current band of current class. Only the non-noise spectra are used in the second training. The resulting prediction coefficients from the second training will be used for prediction and sent to the decoder. The two-times trainings remove part of the interference of noise spectra, and reaches a better compression effect than other methods based on regression prediction.

Image mosaic is brought up for improving the visual field of common camera which cannot fulfill practical requirement of ultra-wide-angle panoramic images. As widely known, to get mosaic images, two main steps should be considered, namely image alignment and stitching. Image stitching algorithms create the high-resolution photo-mosaics used to produce today’s digital maps and satellite photos. In this paper, an image stitching process to detect and eliminate the pixel merges is improved and compared with some popular image stitching algorithms.

Due to long term over-exploring groundwater, ground subsidence has taken place in Urumqi city for many years. Traditional ways of monitoring ground deformation utilize levelling and global positioning system (GPS) measurement. They have the advantage of high accuracy. However, they are very costly and cannot achieve enough spatial sampling density. Recently, space-born synthetic aperture radar interferometry (InSAR) is playing an important role in monitoring ground deformation. In this paper, 11 ALOS PALSAR images from 2007 to 2010 have been acquired to monitor the Urumqi City using small baseline time series InSAR technique. Results show that the subsidence is mainly taken place in Qidaowan Industry Park, Urumqi Development Zone and North Industry Park. The maximum subsidence velocity can reach to -64.6mm/year.

This paper presents a new image enhancement method based on self-adaptive piecewise linear transformation. It is most important to select division points during the processing course of piecewise linear transformation because these division points can decide the quality and efficiency of image enhancement. In this paper, division points are automatically and self-adaptive selected based on the gray grade and histogram feature of images. In order to increase the contrast of the image, it is necessary to extend the object section, maintain the transition section and compress the background section. Our method is easy to use, and the experiment results show that the technique can produce good results on a variety of images.

Phase diversity speckle (PDS) is an image restoration technique which is based on the idea of phase diversity (PD). It uses multi-frame short-exposure image sequence to calculate their corresponding wave-front information. Each image pair consists of two images collected by two cameras at the same time with one in focus and the other with known defocus value. Multi-frame processing can significantly improve the target signal to noise ratio (SNR), and decrease noise influence. In this paper, based on the principle of pupil Fourier imaging, by adjusting the pupil size, we get different scales of the optical point spread function (PSF). Also, we analysis different camera noise distribution channels, location differences and other factors to improve the processing speed of image restoration. In the indoor environment, we build optical platform, and use multi-frame phase diversity speckle to make experiment under different turbulence conditions. The experimental results show that with the camera size of 128×128 pixels, the image restoration effect is close to the diffraction limit.

This paper presents a new approach to automatic registration of terrestrial laser scanning (TLS) point clouds utilizing a novel robust estimation method by an efficient BaySAC (BAYes SAmpling Consensus). The proposed method directly generates reflectance images from 3D point clouds, and then using SIFT algorithm extracts keypoints to identify corresponding image points. The 3D corresponding points, from which transformation parameters between point clouds are computed, are acquired by mapping the 2D ones onto the point cloud. To remove false accepted correspondences, we implement a conditional sampling method to select the n data points with the highest inlier probabilities as a hypothesis set and update the inlier probabilities of each data point using simplified Bayes’ rule for the purpose of improving the computation efficiency. The prior probability is estimated by the verification of the distance invariance between correspondences. The proposed approach is tested on four data sets acquired by three different scanners. The results show that, comparing with the performance of RANSAC, BaySAC leads to less iterations and cheaper computation cost when the hypothesis set is contaminated with more outliers. The registration results also indicate that, the proposed algorithm can achieve high registration accuracy on all experimental datasets.

Spatial-frequency information of a given image can be extracted by associating the grey-level spatial data with one of the well-known spatial/spatial-frequency distributions. The Wigner-Ville distribution (WVD) has a good characteristic that the images can be represented in spatial/spatial-frequency domains. For intensity and range images of ladar, through the pseudo Wigner-Ville distribution (PWVD) using one or two dimension window, the statistical property of Rényi entropy is studied. We also analyzed the change of Rényi entropy’s statistical property in the ladar intensity and range images when the man-made objects appear. From this foundation, a novel method for generating saliency map based on PWVD and Rényi entropy is proposed. After that, target detection is completed when the saliency map is segmented using a simple and convenient threshold method. For the ladar intensity and range images, experimental results show the proposed method can effectively detect the military vehicles from complex earth background with low false alarm.

Natural images often suffer from common problem of poor resolution and low SNR. However, conventional methods can’t accurately detect the edge for this type of images. In this paper, we present a robust multi-scale edge detection algorithm for noisy images. Firstly, we down-sample the image to multi-scale resolution, and then the edge features are exacted with the improved difference eigenvalue algorithm. Finally, the multi-scale edges are combined to the accurate edge using the method of Improved New Edge Direction Interpolation (INEDI). Experimental results show that the proposed method outperforms the conventional methods while suppressing noise and preserving edge thus is suited for nature images edge detection.

As a new technique used to protect the copyright of digital productions, the digital watermark technique has drawn extensive attention. A digital watermarking algorithm based on discrete wavelet transform (DWT) was presented according to human visual properties in the paper. Then some attack analyses were given. Experimental results show that the watermarking scheme proposed in this paper is invisible and robust to cropping, and also has good robustness to cut , compression , filtering , and noise adding .

This paper presents a new visible watermarking solution for CMOS image sensor which can enhance secure features of captured images. Visible watermarks are embedded in the Bayer format image data and can be transferred by the subsequent interpolation process. A piecewise function is setup based on the gray scale resolution characteristics of human eyes. Watermark stretch factor can be adaptively chosen according to the gray value of the current pixel. The advantage of this algorithm is that the watermark has the same visibility in different image brightness region. A number of color images have been used to test the method. In order to check the robustness of watermarked images, we conducted adding noise and filtering experiments, results show that the visibility of watermark is also good after the experiments. The approach allows a digital watermark to be embedded in an image immediately upon its capture, before leaving the imaging chip.

In the field of remote-sensed image segmentation, it is very important to obtain semantic results. However, in high resolution remote sensed images, different complex patterns always have components with the same spectrum, which makes it rather difficult to extract such patterns only through traditional clustering methods. In this paper, a novel multi-stage region-level clustering method is proposed to solve this problem. Firstly, the initial oversegmentation is obtained by using the Mean Shift algorithm, based on which a region adjacent graph (RAG) is built; Then, FCM is employed to get the spectral-based segmentation result; After that, the context clues for each region is calculated according to the label and size of neighboring regions, followed by the second FCM clustering on each set of regions with the same label to distinct regions with the same spectrum but belongs to different objects; Rearranging all of these clustering results to form the finial processing unit, this algorithm goes a step further to calculate more accurate context clues, and use the third FCM to obtain the final segmentation result. Experiments on the high resolution remote-sensed images have shown the superiority to the competitions.

The color transfer is a simple and efficient method to transfer the color in a daytime image to an infrared image so that the infrared image appears a natural color appearance. In order to make the method more conducive to hardware implementation, in this paper, instead of in lαβ color space, the effect of color transfer in YCbCr color space was studied. Instead of a nonlinear color space conversion by linear one, the color transfer in YCbCr color space is simpler, and experimental results shows that the effects of color appearance is similar with that of color transfer in lαβ color space.

Combining the theory of wavelet transform based image fusion and SOPC design method, the authors uses SOPC as the core device to design and implement a image fusion system. The fusion system adopts the Verilog hardware description language, Dsp builder and Quartus II development platform together with macro module to complete the logic design and timing control of each module. In the fusion system, we can achieve simple pixel-level image fusion of two registered images. This design not only builds up an image fusion system based on SOPC in accident, but also provides a hardware design principle in SoPC for the future design and Implementation of more comprehensive function of image processing.