This PDF file contains the front matter associated with SPIE Proceedings Volume 11136, including the Title Page, Copyright Information, Table of Contents, Author and Conference Committee lists.

We frequently find in medical imaging that patient data are “sparse.” That is, when the recorded data are decomposed into an appropriate basis, the information related to a specific clinical task is found in a small compact subspace. Sampling sparse data appropriately enables high frame-rate imaging with minimal loss of image quality. It also enables efficient implementation of machine-learning and other analysis techniques designed to enhance the diagnostic performance of that modality. These ideas are fundamentally changing how we approach data sampling in image science. We are leveraging the advantages of sparsity in developing a new power-Doppler imaging method using data from commercial ultrasound instruments. Our methods significantly increase blood-signal sensitivity and specificity for slow, spatially disorganized patterns of blood flow as is characteristic of peripheral perfusion. We arrange the recorded echo data into high-dimensional arrays that are decomposed into basis sets to effectively separate strong tissue echo signals from the much weaker blood signals of perfusion. That is, we first expand echo-data dimensionality to capture and then isolate the perfusion subspace before reducing dimensionality to render an image. This combination of pulse sampling and clutter filtering enhances peripheral perfusion images such that injectable contrast media are no longer required. In preclinical mouse studies, we find our methods can significantly enhance the effectiveness of sonography at assessing the time course of revascularization in an ischemic hindlimb. The clinical application we are pursuing is serial assessment of peripheral arterial disease in diabetic patients.

To obtain surreal and richer visual experience, augmented reality (AR) technology has been widely used in various areas. As a popular solution of AR, display using computer generated hologram (CGH) is often accompanied by blurring which is caused by uncontrolled interference. In this paper, a modified algorithm based on double-phase hologram (DPH) algorithm is proposed to reduce speckle noise in holographic reconstruction. The macro-pixels in the original hologram are separated into multiple sub-holograms, and these sub-holograms are displayed alternately in high frequency, which reduces the speckle noise generated from the interference between adjacent macro-pixels. Meanwhile, the method is less time-consuming than the traditional Gerchberg-Saxton algorithm because no iteration is needed. The simulation and the optical experiment based on liquid crystal on silicon (LCoS) have been conducted, and the results confirm the feasibility of the proposed method to improve the image quality.

Evaluating the stress distribution in structures under temporal loads is being carry out by many of the engineering applications such as: impacts, cracks, bending, thermal-transient and other. In those cases, conventional photoelasticity techniques are more complex to evaluate the stress field because of their complicated and expensive experiments, quantity of computational procedures, and their time by time analysis. However, dynamic photoelasticity experiments produce temporal information, such as color variations, which could be analyzed, described, and classified in order to perform a whole stress field evaluation. In this paper, the one-dimensional local binary patterns (1D-LBP) are used to describe such color variations and use them to identify the stress values they belong. For different experimental configurations, this proposal achieved an accuracy of 98% when evaluating the stress field of cases with similar light sources than with a reference experiment, and 92% for experiments with other light conditions. These results make this descriptor able to determine categorical stress maps from a photoelasticity video itself, which significantly opens new opportunities to simplify the experimental and computational operations that limit the stress evaluation process in line with the dynamic experiment.

Emulating the sense of touch is fundamental to endow robotic systems with perception abilities. This work presents an unprecedented mechanoreceptor-like neuromorphic tactile sensor implemented with fiber optic sensing technologies. A robotic gripper was sensorized using soft and flexible tactile sensors based on Fiber Bragg Grating (FBG) transducers and a neuro-bio-inspired model to extract tactile features. The FBGs connected to the neuron model emulated biological mechanoreceptors in encoding tactile information by means of spikes. This conversion of inflowing tactile information into event-based spikes has an advantage of reduced bandwidth requirements to allow communication between sensing and computational subsystems of robots. The outputs of the sensor were converted into spiking on-off events by means of an architecture implemented in a Field Programmable Gate Array (FPGA) and applied to robotic manipulation tasks to evaluate the effectiveness of such information encoding strategy. Different tasks were performed with the objective to grant fine manipulation abilities using the features extracted from the grasped objects (i.e., size and hardness). This is envisioned to be a futuristic sensor technology combining two promising technologies: optical and neuromorphic sensing.

We incorporate the prediction of flicker to a semiphysical and analytical model describing the angular and wavelength dependencies of retardance in parallel aligned liquid crystal (PA-LC) devices. This makes the model unique due to the wide range of calculation it offers while keeping its simplicity. Prediction of the modulation of retardance and its associated flicker relies on the fitting of the equivalent tilt angle of the molecules as a function of applied voltage. Specific results are given for liquid crystal on silicon (PA-LCoS) microdisplays, central to many spatial light modulation applications such as the generation of structured polarized beams. Experimental characterization results at arbitrary angles and wavelengths prove the predictive capability of the model. To highlight the richness of situations with PA-LCoS devices, we provide results for two different digital addressing sequences producing different levels of flicker. We focus on the application of the PA-LCoS as a polarization state generator (PSG) and we emphasize the ability of our approach to evaluate the performance across the visible spectrum and for a wide range of incidence angles. Our approach offers novel capabilities in the generation of arbitrary states of polarization, both fully and partially polarized.

In recent years, three-dimensional (3D) display has received widespread attention. The light field display technology based on multi-layer translucent structures enables observers to directly acquire 3D scenes without wearing any auxiliary equipment, and has the advantages of high resolution and low cost. The use of liquid crystal panels as translucent structures makes it possible to realize dynamic 3D display. However, interactive 3D display can hardly be achieved due to the unacceptable long time to generate each frame of a 3D animation. In this paper, we reduce the time consumption for each frame by optimizing acquisition of the four-dimensional light field and calculation of the multi-layer LCD images. With the help of powerful rendering capabilities of OpenGL, we easily obtain the light field information of 3D scenes within less than 0.5s. The light field is rapidly decomposed into multiple LCD images utilizing parallel computing of graphics processing units. Human-computer interaction is realized through the development of Kinect. A 3D display system based on multi-layer LCDs is built, information flow between various components in the system is created, and interactive 3D display is implemented.

With the development of ultra-high definition television (UHDTV), an effective gamut mapping algorithm (GMA) from high-definition television (HDTV) to UHDTV is vital. In this paper, we propose a hue-preserved GMA in the CIELCH color space. To balance the restriction between the color difference and the utilization rate of gamut, a mapping method based on barycenter transformation is used. With this algorithm, the color signals for HDTV can be applicable to UHDTV and the image quality can be improved significantly.

The capture of panoramic images requires the use of complex and specialized cameras. However, high quality panoramic images can be constructed digitally by stitching several images captured with conventional lowcost cameras. In this work, an image stitching method based on projective transformations is proposed. The theoretical principles and computational implementation are presented. Experimental panoramic images are composed to validate the usefulness of our method.

Uncalibrated camera-projector fringe projection systems are unable to provide metric three-dimensional measurements. The main difficulty for camera-projector calibration is that independent calibration of the devices is cumbersome and susceptible to alignment errors. In this paper, an efficient and accurate method for calibration of a camera-projector pair is proposed. The operating principle and computational implementation are analyzed. The metric measurement of a three-dimensional object is carried out to demonstrate the efficiency and accuracy of the proposed method.

We investigate the feasibility of additively manufacturing optical components to accomplish task-specific classification in a computational imaging device. We report on the design, fabrication, and characterization of a non-traditional optical element that physically realizes an extremely compressed, optimized sensing matrix. The compression is achieved by designing an optical element that only samples the regions of object space most relevant to the classification algorithms, as determined by machine learning algorithms. The design process for the proposed optical element converts the optimal sensing matrix to a refractive surface composed of a minimized set of non-repeating, unique prisms. The optical elements are 3D printed using a Nanoscribe, which uses two-photon polymerization for high-precision printing. We describe the design of several computational imaging prototype elements. We characterize these components, including surface topography, surface roughness, and angle of prism facets of the as-fabricated elements.

Length measurements provide important information about the three-dimensional world. This is especially useful for decision making in robot vision, path planning in autonomous navigation, and people identification in security application. In this work, we present a length measurement method based on perspective transformations using an uncalibrated camera. The theoretical principles are analyzed and the computational implementation is discussed. The usefulness of our proposal is verified experimentally by measuring relative lengths from experimental monocular images.

We proposed a Frenet-Serret descriptor to classify stress categories based on color dynamics of pixels stored in photoelasticity videos. A collection of image compression models of disc and ring with a monotonic incremental load was generated. For each pixel, a temporal curve was created using color changes each frame. A descriptor histogram with Frenet Serret parameters was used to train a neuronal network; it classified in four types of stress zones (concentrated, high, medium, low). With the proposed method, a dynamic differentiation was possible in the field of stress without considering traditional digital photoelastic procedures.

Tsang et al. have shown that the Fisher information of the two incoherent point source separation, below the Rayleigh limit, is finite and achievable using optical modes measurements.¹ However, recent claims regarding partial coherence of sources, no matter how small, leads to necessarily zero Fisher information as the source separation decreases below the Rayleigh limit approaching zero have proved to be controversial.^{2, 3} Thus, the impact of partial coherence on the photon counting optical modal measurements merits further exploration. In this work, we derive the mutual coherence function (image plane) of two partially coherent point sources and find the classical Fisher information of the source separation using both direct image plane and photon counting modal measurements. A classical Fisher information analysis of partially coherent source(s) leads to some rather surprising results for two-point source resolution as the source separation approaches zero. We find that the magnitude of the Fisher information strongly depends on the degree of (positive/negative) partial coherence, which can be understood using an intuitive semi-classical analysis of direct image plane and photon counting modal measurements. We also provide an error analysis of the maximum likelihood estimators for both measurements.

Aiming at the current status that the on-orbit imaging accuracy is low and the application-oriented imaging system is not perfect, the detailed simulation of entire space-based imaging chains of space object is carried out in this paper. Firstly, the influence factors on entire space-based imaging chains of space object are analyzed, including the physical parameters of the object, the performance parameters of the detector, and the orbit parameters of the sun, object and detector. Then, a model for entire space-based imaging chains of space object is built. The model consists of radiation transmission based on the bidirectional reflectance distribution function (BRDF) and image degradation based on the modulation transfer function (MTF). Finally, the grayscale images of the International Space Station (ISS) are simulated. It shows the blurring effect caused by each link of entire chains can be truly, accurately and completely reflected.

As machine-learning algorithms continue to expand their scope and approach more ambiguous goals, they may be required to make decisions based on data that is often incomplete, imprecise, and uncertain. The capabilities of these models must, in turn, evolve to meet the increasingly complex challenges associated with the deployment and integration of intelligent systems into modern society. Historical variability in the performance of traditional machine-learning models in dynamic environments leads to ambiguity of trust in decisions made by such algorithms. Consequently, the objective of this work is to develop a novel computational model that effectively quantifies the reliability of autonomous decision-making algorithms. The approach relies on the implementation of a neural network based reinforcement learning paradigm known as adaptive critic design to model an adaptive decision making process that is regulated by a quantitative measure of risk associated with each possible decision. Specifically, this work expands on the risk-directed exploration strategies of reinforcement learning to obtain quantitative risk factors for an automated object recognition process in the presence of imprecise data. Accordingly, this work addresses the challenge of automated risk quantification based on the confidence of the decision model and the nature of given data. Additionally, further analysis into risk directed policy development for improved object recognition is presented.

Template matching is an effective method for object recognition because it provides high accuracy in location estimation of targets and robustness to the presence of scene noise. These features are useful for vision-based robot navigation assistance where reliable detection and location of scene objects is essential. In this work, the use of advanced template matched filters applied for robot navigation assistance is presented. Several filters are constructed by the optimization of objective performance criteria. These filters are exhaustively evaluated in synthetic and experimental scenes, in terms of efficiency of target detection, the accuracy of a target location, and processing time.

A robust algorithm for Japanese handwritten hiragana character classifier is proposed using a machine learning approach for minimal training data to reduce computational power and time consumption. The proposed algorithm utilizes image recognition techniques to process samples from a data set. Six different models involving convolutional neural networks are implemented using image templates that were previously processed, in order to achieve great results with the least possible amount of training data. Prediction results were evaluated separating the dataset in training and validation data at a ratio of 5:95 respectively, achieving 96.95% as the highest accuracy across different models, competing against state-of-the-art classifiers with 80:20 training ratio.

Traffic sign detection is a crucial task in autonomous driving systems. Due to its importance, several techniques have been used to solve this problem. In this work, the three more common approaches are evaluated. The first approach uses a model of the traffic sign which is based in color and shape. The second one enhances the image model of the first approach using K-means for color clustering. The last approach uses convolutional neural networks designed for image detection. The LISA Traffic Sign Dataset was used which it was divided into three superclasses: prohibition, mandatory, and warning signs. The evaluation was done using objective metrics used in the state-of-the-art.

Convolutional neural networks (CNNs) is becoming a critical role for deep learning-based computer vision applications. Through CNN we can extract meaningful information out of massive sensor data. Vision applications use this information to analyze the mainstream trends of the data and take immediate action based on these trends. However, CNN's energy consumption and bandwidth limitations make it difficult for CNN network systems to deploy in mobile systems with stringent energy limits. In this paper, we explore the simulation and the hardware method of optical convolution for low power image processing, which is inspired by the bio-image sensor Angle sensitive pixel (ASP). This optical computation method may be used to substitute the first convolution layer of the CNN network due to its energy-saving features and speed of light processing time. We adopt two-layer of customized transmission grating to perform this optical convolution computing. By generating the Talbot effect, the two-layer grating structure can perform optical convolution computation using Gabor wavelet filters, which will cause zero electrical power. We demonstrate both simulation and experiment results for optical convolution through our algorithm and prototype system, the convolution results can extract different meaningful information about the original image, which is very similar to edge filtering. This optical operation will hopefully be used to replace the first convolution layer of CNN since it can effectively reduce both the consumption of the energy power and the performing time.

This paper proposes frequency-domain correlation filtering to solve object recognition of three-dimensional (3D) targets. We perform a linear correlation in the frequency domain between an input frame of the video sequence and a designed filter. This operation measures the correspondence between the two signals. In order to produce a high matching score, we design a bank of correlation filters, in which each filter contains unique information of the target in a single view and statistical parameters of the scene. In this paper, we demonstrate the feasibility of correlation filters used to solve 3D object recognition and their robustness to different image conditions such as noise, cluttered background, and geometrical distortions of the target. The evaluation performance presents a high accuracy in terms of quantitative metrics.

Detection and description of local features in images is an essential task in robot vision. This task allows to identify and uniquely specify stable and invariant regions in a observed scene. Many successful detectors and descriptors have been proposed. However, the proper combination of a detector and a descriptor is not trivial because there is a trade-off among different performance criteria. This work presents a comparative study of successful image feature detection and description methods in the context of the simultaneous localization and mapping problem. The considered methods are exhaustively evaluated in terms of accuracy, robustness, and processing time.

This paper presents a proposed algorithm with the implementation of the A* algorithm for path planning in a partially known environment. By using a differential mobile robot, the navigation is accomplished with a LiDAR sensor that detects any potential changes in the environment. The proposed algorithm estimates a safety path-planning trajectory from the origin of the robot to a target coordinate given by the user. If the robot encounters an unknown obstacle that does not belong to the known environment it will update the map, and recalculate the trajectory, executing it and proceed with the new path. Experimental results were considered in an indoors cluttered environment given by unknown obstacles, and partially known maps.

The blood oxygen saturation (SpO2) detection is of great importance in medical science due to its relationship with cardiopulmonary function. Determination of SpO2 by infrared spectroscopy is a commonly used method. In this paper, a novel non-invasive blood oxygen saturation detection method with video image is proposed. Firstly, the index fingertip is illuminated by LEDs with four wavelengths. Then, the reflected light is collected by an image sensor. Meanwhile, a finger clip pulse oximeter is adopted to monitor SpO2 changes. Finally, a color regression model is established with the obtained RGB values and the corresponding SpO2. Furthermore, the influence of quantization noise and other factors is analyzed. Experiments with multiple samples are carried out using the proposed method. The results show that the relative errors between the predicted and reference values are within 5%. Compared with traditional methods, the proposed method can effectively detect the SpO2 in a specific area instead of a single point. In addition, it provides an alternative approach that guides an SpO2 detection device for daily use.

Chaos has the deterministic and unpredictable aspects useful for secure communications. Chaos encrypted optical signals provide high degree of security in free space as well as in optical fibers, but the effects of propagation through these mediums were not explored. In this paper, the evolution of Lyapunov exponent of optical chaos encrypted signal in optical fibers is studied. To study the effects of fiber channel, three different chaotic optical encryption schemes are studied and compared.

We present a Forward Error Correction (FEC) assisted optical link that operated at 28 Gbaud (56 Gbps) error-free over link lengths ranging from 10 m to 40 km with Pulse Amplitude Modulation 4-level (PAM4) signals. The links operated with link margins greater than 10 dB due to the wide optical dynamic range performance which will be evident from the sensitivity plots presented. The link was implemented with a commercial PAM4 transceiver chip with built-in FEC driving a Low Noise Amplifier (LNA) which in turn drove a 1550 nm Externally Modulated Laser (EML). The link could have been implemented with a 1310 nm laser and produced comparable results. The optical signal was transmitted over dispersion compensated fiber, fed to a low-noise 28 GHz Linear InGaAs photoreceiver having variable gain, and routed back to the PAM4 transceiver chip. Real time bit error rate measurements demonstrate the advantage of employing linear receivers having Automatic Gain Control (AGC).

We propose a screening system based on nano structure photonics crystal in homecare. Proposed system used several kinds of photonics crystal based substrates: 50nm L/S SERS substrate, 70nm L/S SERS substrate, 100nm L/S SERS substrate. Target drug in tear drop on these fabricated substrates. We confirmed the effect using Surface enhanced Raman scattering. In this study, Target drug is Phenobarbital Sodium (PB) which is an anticonvulsant agent. The results suggest a form that could enhance Raman Spectroscopy efficiently.

A real-time system for restoration of images degraded by haze is presented. First, a transmission function estimator is automatically constructed using genetic programming. Next, the resultant estimator is employed to compute the transmission function of the scene by processing an input hazy image. Finally, the estimated transmission function and the hazy image are used in a restoration model based on atmospheric optics to obtain a haze-free image. The proposed method is implemented in a laboratory prototype for high-rate image processing. The performance of the proposed approach is evaluated in terms of objective metrics using synthetic and real-world images.

Fast beam steering is useful in applications including free space optical switching and communications. Quasi-static beam steering, where the beam rapidly switches between discrete fixed directions, presents special challenges. Here we demonstrate a fast quasi-static ‘pinwheel’ scanner by conformally mapping linear blazed gratings into curved structures fabricated onto annular sections of a rotary disk. We use Matlab and Zemax to model the effects of the conformallymapped grating on the emitted optical beam. We show a specific two-dimensional (2D) ‘pinwheel’ scanner design with 56 gratings, each deflecting 1.31 μm incident light by 11.3° in one of four directions with 75% optical efficiency. The element was fabricated by optical grey-scale lithography on a 95 mm diameter substrate, coated with gold, and mounted onto the spindle of a 3.5” format 7200 rpm magnetic disk drive. We characterize the optical beam steering efficiency, pointing, and stability, and demonstrate microsecond switching speed of a single mode fiber signal.

Chaotic optical signals may be generated by using an acousto-optic system in the Raman-Nath mode. The bistability in the Raman-Nath mode is realized by detecting the intensity of the diffracted optical beams and feeding back the electronic signal to drive the acousto-optic cell. If the parameters of the system are chosen ”judiciously”, the intensity of the output beams oscillates in a chaotic manner in time. The conditions for generating chaos from various diffraction orders of the Raman-Nath cell are derived by calculating the Lyapunov exponents. If the Lyapunov exponent is a positive number then the system generates chaos; otherwise the system remains stable and produces a well-defined output signal. When the Lyapunov exponent for a RamanNath diffraction order is known the physical parameters are calculated so that chaos is produced in the output of the system in that diffraction order.

Nowadays, with the boosting incomes and rapid development of science, car has become one of the most important transport vehicle. Driving environment and safety has attracted much attention in the automotive design field. Therefore, it is extremely urgent to develop the intelligent and reliable safety technologies such as vehicle active collision warning system. There are lots of studies focused on the critical research of machine vision ranging technology, however, the installation error of binocular ranging may result in inaccurate measurement accuracy. In this study, we use an improved monocular ranging to measure distance. The traditional monocular ranging models based on the principle of pinhole imaging, static image ranging model and etc. Most of the models require specific prior information of the vehicle, the applicable conditions of the model may be too idealistic and not applicable to general situations. In order to solve the contradiction, our research proposes a creative monocular ranging model to measure the distance between two vehicles. The model is based on the camera space projection relationship taking the factors of the camera's pitch angle into account. Our model has universal application significance using the simple implementation method with residual method. Based on the model, we amended the camera's pitch angle after the experiments. Meanwhile, the accuracy of the model is guaranteed by analyzing the factors affecting the accuracy of the range. The experimental results show that the error is controlled within 10%, which can meet the accuracy requirements of the system.

Camera’s internal parameter calibration is an important problem in computer vision. The moment, a precise targets like checkboard is required when camera calibration is performed. But this method is not suitable for a telephoto lens. In order to overcome this problem, a new camera calibration model is constructed by installing the debugged telephoto camera on an accurate two-dimensional rotating platform. Let the camera move around the rotating axis of the rotating platform. The camera takes pictures of distant objects directly. Then, the next picture is get by rotating the two-dimensional platform. The image coordinates , of the same point in space, in different images are obtained by image matching. The motion matrix of the camera is calculated from the readings of the two-dimensional rotating platform. Last, the optimization equation can be enumerated to solve the internal parameters. The experimental results show that the internal parameters obtained by the algorithm can satisfy the calculation accuracy.

This paper is devoted to study of speckle field dynamics during coherent light scattering by polycrystalline structure surface in the process of consolidation. Accent is made on polycrystalline material which includes carbon nanostructures. Such inclusion promises significant increase of mechanical properties. Analyses of speckle field dynamic shows possibility for identification of consolidation processes which happens in polycrystalline structure during formation. Comparison between speckle fields dynamic formed by polycrystalline structure with and without carbon nanostructures demonstrates ability of their differentiation. Paper shows that study of intensity fluctuations of scattered coherent radiation is suitable technique for the analysis of polycrystalline structure consolidation.

Parallel-aligned liquid crystal on silicon (PA-LCoS) microdisplays are widely used in spatial light modulation applications, especially in those requiring phase-only modulation. One such application area is programmable diffractive optics which plays a very important role in modern optical imaging systems or in optical interconnections for optical telecommunications. Among the multilevel diffractive optical elements (DOEs) we focus on the important case of the blazed gratings. We develop the corresponding analytical expressions for the diffracted field where, as one of the novelties in the work, an analytical expression including the fill factor and the flicker is obtained. This enables to have a model against to compare the experimental results in a number of situations where fill factor, flicker, period, and number of quantization levels are the variables. This also enables to design appropriate compensation techniques to enhance the performance of the blazed gratings.

Scattering media would scratch light propagation, and images would degenerate into unrecognizable speckle patterns. Conventional target recognition through scattering media is composed of two steps, i.e., reconstruction and recognition. Here, combining the compressive sensing with feature extraction, a method of efficient speckle based compressive target recognition through scattering media is proposed. In the paper, autocorrelation of speckles is proved to have the same singular values as that of their corresponding objects, and then speckle based recognition is introduced. Compressive sensing can be used to retrieve signals with measurements fewer than those required by Nyquist-Shannon theory. With the proposed method, scattered object recognition can be replaced with speckle recognition, bypassing the conventional object reconstruction procedure. Performances are validated through relevant experiments. Besides, benefited from the conclusion, domain adaption based support vector regression method is proposed and utilized for imaging through scattering media then. Domain adaption is introduced to transfer leaning samples and testing samples into a new space where the distance between them is much closer, leading to high reconstruction fidelity in the followed support vector regression based inverse scattering stage. Principle component analysis is also considered to help decrease dimension and thus improving efficiency. Experiments validate that the presented technique owns a higher image reconstruction efficiency and fidelity, compared with our previous researches. Since the target recognition and reconstruction is mainly based on ground truth images, the work is valuable and meaningful for remote sensing applications, especially for object detection or monitoring when scattering is occurred.