KEYWORDS: 3D modeling, Calibration, Sensors, Covariance matrices, Pose estimation, Point clouds, 3D image reconstruction, Data modeling, Reconstruction algorithms, RGB color model
The 3D reconstruction with multiple RGBD sensors is favored for its excellent real-time performance and high accuracy, where the performance largely depends on the precision of calibration among RGBD sensors. This paper proposes a graph-optimization-based calibration method for multiple RGBD sensors. Initially, a coarse calibration is assisted by the Zhang's calibration method; based on this, a fine calibration is further conducted through ColorICP to correct the extrinsic parameter estimation between each pair of sensors. Concurrently, this method computes the information matrix of the extrinsic parameter estimation. Utilizing both the extrinsic parameter estimation and the information matrix, a pose graph is constructed for global optimization estimation. The efficacy of this scheme in enhancing the accuracy and robustness of multi-RGBD calibration is demonstrated through applications in a gazebo simulation environment and a physical 3D reconstruction system built with four Intel Realsense L515 cameras, showcasing the potential applications of multi-RGBD sensor 3D reconstruction technology.
The principle of laser triangulation has been widely used in various high-precision measurement due to its advantages of high accuracy, fast response speed and simple structure. However, the laser triangulation method still has some limitations for the measurement of dynamic object, and the accuracy of laser location will be significantly reduced especially when the working distance becomes longer. We have carried out the development and performance optimization of a long-distance laser triangulation displacement sensor for the application of dynamic object. The measurement uncertainty is analyzed when the laser dithering, laser drifting and object motion are all taken into account. Then, reflectors are added to the optical path structure of traditional laser triangulation system with the aim to improving its sensitivity and expanding its working range. The experiment results show that the repeatability of our designed optical system is 2.9μm and the imaging uniformity is 86.23% when the working distance is of 1300mm.
Water surface roughness is important but challenging to measure parameter which reflects the energy transfer at the air-water interface. This paper presents a field experiment obtaining lake surface video images (Fig. 1). Also, it establishes the relationship between the aerodynamic roughness of the lake surface (or wind speed) and the characteristics of the lake surface video images based on texture features and fractal dimensions. This work is a preliminary study of sea surface roughness measurement. The texture features and fractal dimensions are calculated by using the methods of gray level co-occurrence matrix, gray level-gradient co-occurrence matrix, autocorrelation function, Tamura texture feature, fractional Brownian motion autocorrelation, box counting, improved box counting, gray statistical increment, gray statistical definition and area measurement. The empirical values of lake surface roughness are found from measured wind speed and an empirical relation. The correlations between lake surface roughness (or wind speed) and texture features (or fractal dimensions) are evaluated based on the data from the field experiment. Furthermore, three types of noises with different parameters are introduced to the lake surface video images. Then noise suppression performances of these methods are evaluated. The experiments have demonstrated that lake surface image roughness calculated by texture (or fractal) methods and empirical relation between wind speed and lake surface roughness is effective for analyzing lake surface roughness. The running time of various methods is calculated to analyze the possibility of real-time detection. Plans for further investigation of lake or sea surface roughness features are also discussed.
Laser triangular displacement sensor (LTDS) is widely used in various high-precision measurement scenarios, but the accuracy of the sensor will be distinctly reduced in a working environment with extreme temperature. The mechanical structure, optical crystal and illuminant of the sensor will be distorted or drifted due to the impact of temperature change. This paper focuses on the spatial drift characteristics of illuminant on the measured surface with temperature change. The self-developed LTDS prototype is used to conduct the temperature-rising experiment of the "illuminant - collimated lens" system, and the centroid drift of the spot formed by the illuminant under two different power supply currents (16mA and 24mA) is recorded. The experimental results show that the centroid of laser spot appears clustering phenomenon in the process of heating. After filtering the noise generated by clustering phenomenon, the drift of the centroid of laser spot in the spatial plane presents a linear directional law, and the drift distance presents a quadratic function relationship with the change of temperature. In addition, the error evaluation model is used to calculate the actual measurement error of the sensor caused by the centroid drift of the spot. The measurement error of the sensor is quadratic function with the change of temperature. Comparing the results of heating experiment under 16 mA and 24 mA power supply, choosing appropriate power supply current (24 mA) can effectively suppress clustering phenomenon and improve the stability of illuminant.
An irradiance distribution model for laser triangulation displacement sensors is proposed, which is more suitable for practical applications than the geometric optical model (GOM). Based on the irradiance distribution model, a convergence algorithm is developed. Compared with the design parameter optimized method based on the GOM, the convergence algorithm can be used for optimization of not only the traditional design parameters including view angle and object distance but also the waist of the source laser beam. Additionally, using this convergence algorithm, the nonlinearity can be evaluated in the design stage. To validate the convergence algorithm, two example cases are performed. The results show that the nonlinearities are reduced to 6.3098 and 5.7946 μm, respectively.
Linear laser scanner has been widely used in industrial applications, such as 3D reconstruction. Breakpoint detection of the laser stripe centerline is a fundamental step. In practical applications, the centerline is always not perfect due to various noises. A novel breakpoint detection method named Fixed Dynamic Programming (FDP) is proposed. Firstly, the centerline fitting error is illustrated in detail, which is to be used as the criterion in the FDP. Secondly, principle of the FDP method is described, which modified the traditional DP method by utilizing the foreknown breakpoints information of the centerline. Finally, experiment is implemented to detect breakpoints of the noisy laser stripe centerline by using a linear laser scanning system. From the experiment result verifies that the prosed FDP method can avoid the influence of noise and improve the computation efficiency.
KEYWORDS: Cameras, Facial recognition systems, Sensors, Detection and tracking algorithms, Information fusion, Image fusion, Neural networks, Data fusion, Signal attenuation, Imaging systems
Multi-camera multi-object tracking problem shares the difficulties of both multi-camera information fusion and multiobject data association across time. Top view enjoys nice properties for object detection and tracking, such as information completeness and no occlusion. This paper utilizes the properties to address this problem by transferring object representation to a ‘top view’ from multiple partial views, hence converts multi-camera tracking problem into a singlecamera one. To this end, a detector was applied on each single partial view image firstly. After that, to obtain the representation of these detection resultsin a common top view, we first introduce the ‘real top view supervised transferring’ method, in which, a network was used to transfer the detection-bounding box in partial view to real top view. However, this method depends on top view image as supervision signal during training. We investigate further to eliminate the dependency, so here came the virtual top view, which is in essence a hyperspace. A convolutional network and triplet loss were used to map an object with its position in each partial view to a vector in the hyperspace and supervises the learning of representation respectively. Finally, by applying cluster algorithm on the transferred representation of object in the top view, the information from multiple partial cameras was fused and unified. Tracking on the top view can be formulated as a data association problem, which can be solved by traditional assignment algorithms. Experiment results on our own dataset and EPFL dataset[1] showed the effectiveness of the proposed methods.
Laser triangulation displacement sensor using the single point laser or linear laser as the source is widely used in industry. The sensor uses the triangulation principle among light source, tested object and receiver with high speed and non-contact advantages. However, the structure of most of traditional laser triangulation sensors is split that an external amplifier controller is necessary, causing a complicated system. An integrated structure of laser triangulation displacement sensor is put forward in this paper. All signal collection and processing systems are integrated in the probe so that the sensor can be convenient to measure. The limited volume of sensor leads to the change of layout. So the image quality of spots also changes as well. The layout and the principle of the sensor are introduced. The expression of the light spot is also analyzed. A new measurement method of spot processing is also proposed at the same time. Some experimental results are also given to show the efficiency of the new algorithm.
Camera calibration using a 2D planar board is widely applied because of the flexibility and simplicity of this method. However, this technique fails to yield reliable and accurate calibration results when an imperfect planar target is used. The accuracy of the image and world coordinates for the extracted corners, which are prerequisites for deriving precise camera parameters, are affected by the lens distortion and the non-planarity of the calibration target. In this paper, the accuracy of the obtained image coordinates is improved by combining a Hilbert transform with a traditional calibration method. In addition, the geometry of the calibration target is fully considered so as to acquire precise world coordinates using an overall nonlinear parameter optimization algorithm. The reprojection error of the proposed method is reduced by 80% compared with the traditional method for a significantly deformed planar target, which demonstrates the superiority of the proposed camera-calibration technique.
Fringe projection three-dimensional measurement is widely applied in a wide range of industrial application. The traditional fringe projection system has the disadvantages of high expense, big size, and complicated calibration requirements. In this paper we introduce a low-cost and portable realization on three-dimensional measurement with Pico projector. It has the advantages of low cost, compact physical size, and flexible configuration. For the proposed fringe projection system, there is no restriction to camera and projector’s relative alignment on parallelism and perpendicularity for installation. Moreover, plane-based calibration method is adopted in this paper that avoids critical requirements on calibration system such as additional gauge block or precise linear z stage. What is more, error sources existing in the proposed system are introduced in this paper. The experimental results demonstrate the feasibility of the proposed low cost and portable fringe projection system.
In biological cells and tissues environment, real-time monitoring and controlling dissolved oxygen (DO) provides critical information for studying cellular metabolism process, health status and pathological features. This paper developed an optical DO sensor based on fluorescence quenching principle, prepared tris(4,7-diphenyl-1,10- phenanthroline)ruthenium(II) dichloride complex sol-gel sensing film, and studied its sensing performance. The principle of this sensor is that dissolved oxygen has quenching effect towards the fluorescence emitted by ruthenium complex. So the fluorescence intensity is reduced due to the existence of DO. The measurement limit of DO was 10- 100%, the response time was 20s, and the resolution was 0.02. Compared to traditional dissolved oxygen electrode probe, this luminescent fiber had many advantages, such as smaller size, shorter response time and higher stability.
Large frames made of compound material are widely used in many areas such as industry and aerospace. To ensure the frames working properly, monitor of its deformation after expansion is of great importance. A real time measurement system based on retro-reflective technique and vision method is proposed. Several round retro-reflective marks are placed on some defined positions of the frame as mark points which can precisely describe the shape of the frame. The marks are illuminated by laser beams and the image of the marks is captured by a camera. The positions of the marks on the camera are determined by designed image process program. Experimental results show that the whole process can be finished within 0.1 second. The designed system can measure the deformation of up to 50mm and the accuracy of the measurement is better than 0.02 mm.
A new geometrical model and mathematical method for three-dimensional surface reconstruction with phase- shifting structured light technique is proposed. First, with the proposed method, there will be no restriction to camera and projector's relative alignment on parallelism and perpendicularity. Furthermore, the depth map can be directly derived from the phase map by a simple function. The x and y coordinates are inferred from the depth z coordinate. Second, no special calibration hardware is needed; a simple two-step plane-based calibration procedure is adopted to acquire the system intrinsic and extrinsic parameters. Third, a look-up-table compensation method is presented to avoid the accumulation of errors introduced in the structured light system. The experimental results demonstrate the e®ectiveness and accuracy of the proposed methods.
It is very important to monitor and control pH during cell and tissue culture. On-line pH monitoring provides valuable information on cell metabolic processes and living environment. A novel simple method to real-time measure pH during cell and tissue culture has been experimentally demonstrated using a tapered optical fiber coated with polyaniline. The fiber is tapered to produce the leaking mode in the sensing film. The absorption coefficient and the refractive index of the polyaniline film will vary with different pH values and resultantly change the optical spectral responses. The optical power droped with the increase of the pH at 1042nm. Such a device is sensitive to pH allowing the determination of pH values ranging from 5 to 11 and the resolution of the order of 0.03. Comparing to the conventional pH glass electrode, this optical measurement has smaller size, faster response and can avoid the contamination of the cell and tissue culture fluid.
The pH value has been one of the keen parameters in the cell culture. And the optical measurement method especially the optical method using optical fiber has been widely used due to its small volume and high accuracy. In this work we design a phase detection system to detect the phase difference between the signal and reference signal in the Dual Lifetime Referencing Method. The modulation frequency of the DLR method ranges from 10 kHz to 60 kHz and the system we designed can detect phase difference from the frequency of 1 kHz to 100 kHz. The error can be as small as 0.2%.
As novel sensors, rotational symmetric triangulation sensors (RSTSs) have broad application potential because of their robustness and accuracy. However, existing RSTSs have the disadvantages of complicated structure and high cost. The current paper presents a novel RSTS based on an object space mirror. The proposed sensor uses a cone mirror and a commercial camera, which significantly simplifies the structure and reduces its cost. Mathematical models are discussed to reveal the measurement range of the sensor and the relation between the displacement of the spot and the radius of the ring in the image. An improved image processing method is also introduced for a more efficient processing. Finally, two experimental systems with mirror angles at π/6 and π/4 are set up. Experimental results showed that the proposed sensor can achieve a comparable accuracy of 1.6 μm compared with that of existing methods, as well as a higher efficiency of image processing, while remaining at a low cost.
A new method of arbitrary angle measurement with a grating wedge-plate interferometer is described here. A grating wedge-plate is introduced to the dual-frequency laser interference angle measurement system to enlarge the angle measurement range from ±10 to 360° or more in this system. The optical path and the structure of the interferometer are introduced and its advantages are analyzed. The calibration of the system and absolute zero-point determination of dynamic measurement by relative measurement method using 360° natural angle standard is also provided. Experiments of the repeatability of arbitrary angle measurement without and with self-calibration, the calibration error and the repeatability of zero-point of dynamic measurement under different rotating rates are given. The angle measurement range of the system is up to 360° and the measurement accuracy of arbitrary angles is better than 1” and can reach 0.5” for the best. The largest calibration error of the calibration curve is 0.5” and the repeatability of zero-point in dynamic measurement is related to the rotating rate of the wedge-plate for which the best is 0.3”. High precision, stability, immune to errors and self-calibration are the distinguished features of this angle interferometer.
A new method of rotating angle measurement with dual-frequency laser interferometry using a grating wedge-plate is described here. A grating wedge-plate is formed by making a reflecting flare grating on a glass wedge-plate. The principle of the grating wedge-plate is introduced and its application in the rotating angle measurement system is provided. The setup and principle of the system are also introduced. This system can be applied not only to small angle measurement but also to dynamic random angle measurement. An experiment of rotating angle measurement is given. The measuring range of angle is up to 360 degree(s) and the measurement accuracy is better than 0.5 arc sec. High precision, stability and immune to errors are the distinguished features of this interferometer.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.