- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
[IEEE 2018 25th IEEE International Conference on Image Processing (ICIP) - Athens, Greece (2018.10.7-2018.10.10)] 2018 25th IEEE International Conference on Image Processing (ICIP) - Robust Facial Pose Estimation Using Landmark Selection Method for Binocular Stereo Vision
摘要： In this paper, we present a robust framework for facial pose estimation from binocular stereoscopic vision. Unlike prior work on the facial pose estimation that employs the whole landmarks even located in the wrong position, we propose a landmark selection method to remove the erroneous landmarks for better performance, especially in the large facial pose case. For this purpose, we train a convolutional neural network (CNN) in order to measure the con?dence of each facial landmark detected by using a well-known landmark detection algorithm. Also, by ?tting selected landmarks to 3D space, our framework becomes more robust even when a small number of landmarks are selected. Due to the absence of public dataset for the binocular stereo facial pose, we construct facial pose datasets using a motion sensor for performance validation. In our experiments, our method achieves the higher accuracy of the pose estimation than the previous method, especially for large facial pose cases.
关键词： Binocular stereo vision,Facial landmarks,Facial pose estimation,Stereo matching algorithm
更新于2025-09-23 15:21:01
-
[IEEE 2019 IEEE International Conference on RFID Technology and Applications (RFID-TA) - Pisa, Italy (2019.9.25-2019.9.27)] 2019 IEEE International Conference on RFID Technology and Applications (RFID-TA) - A Waveguide Transmitarray Antenna for Near-field Millimeter-wave Sensing Applications
摘要： Miniaturized hyperspectral imaging sensors are becoming available to small unmanned airborne vehicle (UAV) platforms. Imaging concepts based on frame format offer an attractive alternative to conventional hyperspectral pushbroom scanners because they enable enhanced processing and interpretation potential by allowing for acquisition of the 3-D geometry of the object and multiple object views together with the hyperspectral reflectance signatures. The objective of this investigation was to study the performance of novel visible and near-infrared (VNIR) and short-wave infrared (SWIR) hyperspectral frame cameras based on a tunable Fabry–Pérot interferometer (FPI) in measuring a 3-D digital surface model and the surface moisture of a peat production area. UAV image blocks were captured with ground sample distances (GSDs) of 15, 9.5, and 2.5 cm with the SWIR, VNIR, and consumer RGB cameras, respectively. Georeferencing showed consistent behavior, with accuracy levels better than GSD for the FPI cameras. The best accuracy in moisture estimation was obtained when using the reflectance difference of the SWIR band at 1246 nm and of the VNIR band at 859 nm, which gave a root mean square error (rmse) of 5.21 pp (pp is the mass fraction in percentage points) and a normalized rmse of 7.61%. The results are encouraging, indicating that UAV-based remote sensing could significantly improve the efficiency and environmental safety aspects of peat production.
关键词： radiometry,image classification,geographic information system,spectroscopy,Calibration,geometry,stereo vision,remotely piloted aircraft,remote sensing
更新于2025-09-23 15:19:57
-
An Optical Fiber Immunosensor with a Low Detection Limit Based on Plasmon Coupling Enhancement
摘要： Miniaturized hyperspectral imaging sensors are becoming available to small unmanned airborne vehicle (UAV) platforms. Imaging concepts based on frame format offer an attractive alternative to conventional hyperspectral pushbroom scanners because they enable enhanced processing and interpretation potential by allowing for acquisition of the 3-D geometry of the object and multiple object views together with the hyperspectral reflectance signatures. The objective of this investigation was to study the performance of novel visible and near-infrared (VNIR) and short-wave infrared (SWIR) hyperspectral frame cameras based on a tunable Fabry–Pérot interferometer (FPI) in measuring a 3-D digital surface model and the surface moisture of a peat production area. UAV image blocks were captured with ground sample distances (GSDs) of 15, 9.5, and 2.5 cm with the SWIR, VNIR, and consumer RGB cameras, respectively. Georeferencing showed consistent behavior, with accuracy levels better than GSD for the FPI cameras. The best accuracy in moisture estimation was obtained when using the reflectance difference of the SWIR band at 1246 nm and of the VNIR band at 859 nm, which gave a root mean square error (rmse) of 5.21 pp (pp is the mass fraction in percentage points) and a normalized rmse of 7.61%. The results are encouraging, indicating that UAV-based remote sensing could significantly improve the efficiency and environmental safety aspects of peat production.
关键词： Calibration,radiometry,geometry,image classification,stereo vision,remote sensing,remotely piloted aircraft,spectroscopy,geographic information system
更新于2025-09-23 15:19:57
-
[IEEE 2018 15th International Conference on Control, Automation, Robotics and Vision (ICARCV) - Singapore, Singapore (2018.11.18-2018.11.21)] 2018 15th International Conference on Control, Automation, Robotics and Vision (ICARCV) - Virtual Commissioning of Machine Vision Applications in Aero Engine Manufacturing
摘要： New aero engine design puts new demands on the manufacturing methods with increased automation level. Therefore, the use of vision sensors for control and guiding of industrial robots is being increasingly used. In such system, it is need to customise the machine vision system with real components in the real environment which is normally done close to the start-up of the production. This paper addresses a new concept for designing, programming, analysing, testing and verifying a machine vision application early in the design phase, called Virtual Machine Vision. It is based on a robot simulation software where the real machine vision application is simulated before the implementation in the production line. To verify the Virtual Machine Vision concept an advanced stereo vision application was used. Using two captured images from the robot simulated environment, camera calibration, image analysis and stereo vision algorithms are applied to determine a desired welding joint. The information of the weld joint, i.e. robot position and orientation for the weld path, are sent from the machine vision system to the robot control system in the simulation environment and the weld path is updated. The validation of the Virtual Machine Vision concept using the stereo vision application is promising for industrial use, and it is emphasised that the same programs are used in the virtual and real word.
关键词： simulation,robotics,welding,machine vision,image processing,Virtual Machine Vision,stereo vision,camera calibration
更新于2025-09-19 17:15:36
-
High-speed 3D shape measurement using Fourier transform and stereo vision
摘要： Background: In the fast 3D shape measurement, it is an important factor to use the least number of fringe patterns to get the wrapped phase and the wrapped phase is always required to be retrieved to absolute phase. But the process of phase unwrapping may affect the quality of absolute phase. Besides absolute phase retrieval is time-costing especially for high-speed 3D shape measurement. Methods: This paper proposes a fast measurement method based on Fourier transform profilometry and stereo vision. Because of the property of stereo vision, every point can find its corresponding point in the wrapped phase. This method can reconstruct 3D surface without phase unwrapping. In order to further increase the measurement speed and overcome the gamma effect of projector, the dithering fringe pattern is used. To resist false matching, the original image matching constraint gives a rough parallax. Phase matching and sub-pixel parallax optimization are used to reduce the matching errors. In order to confirm the phase value of points which are at the edge of wrapped phase, average phase value is calculated. Results: A white mask is measured based on the proposed method. The results from every step show the effect of different functions. To better show the effect of the proposed method, a white house is measured and the measurement is compared with LSSM and MFH. In order to further to show the advantage of the proposed method, discontinuous blocks and continuous blocks are measured. The average height, RMS, average error, maximum error are compared with the method which uses absolute phase to match the corresponding points. Experiment results verify the feasibility of the proposed method and it can measure complex objects with high speed and accuracy. Conclusion: This paper uses Fourier transform and stereo vision to get wrapped phase and reconstruct 3D shape without phase unwrapping. The proposed method contains three part: phase matching, edge points detection and sub-pixel parallax optimization. By comparing matching precision, the proposed method can get high quality surface. Because the process of measurement only need one frame of deformed fringe pattern to get the wrapped phase and the matching process does not need phase unwrapping, the proposed method has the potential to be used in fast measurement.
关键词： Fourier transform,Three-dimensional shape measurement,Stereo vision,Fringe analysis
更新于2025-09-19 17:15:36
-
Volume Measurement of Deposits Based on Stereo Vision and SURF
摘要： Based on the stereo vision principle, a volume measurement method of deposit is realized. The principle can be applied for the measurement of deposits with messy texture, such as rubbish, construction waste, and small mounds. After camera calibration and image rectification, the disparity of key points are calculated using SURF feature point matching, and the mismatched point pairs are kick out by the epipolar line constrain. Delaunay triangular meshes are created on the final point pairs. Then the volumes of deposit heap are calculated according to the pinhole imaging model. Experiments show that, despite the use of ordinary USB camera, the result is still relatively accurate. If cameras with high quality and high definition are used, the method can be applied to practical applications.
关键词： SURF,Stereo Vision,Deposits,Volume Measurement
更新于2025-09-19 17:15:36
-
[IEEE 2019 International Conference on Image and Vision Computing New Zealand (IVCNZ) - Dunedin, New Zealand (2019.12.2-2019.12.4)] 2019 International Conference on Image and Vision Computing New Zealand (IVCNZ) - Accurate 3D Measurement of Highly Specular Surface using Laser and Stereo Reconstruction
摘要： In the production line, Inspection and Quality Control are essential to maintain the quality of the products. Same as the manufacturing process, the process of quality checking should also be automatic [1]. In addition to automation in quality checking, the process needs to be simple, fast, accurate and cost-effective. ”Machine vision is the technology and methods used to provide imaging-based automatic inspection and analysis [2].” However, Re?ective Surface of the object puts some limitations on traditional methods of machine vision. This research paper focuses on the development of a simple and accurate 3D measuring system for the inspection of re?ective surface objects. The idea is to combine the concepts of stereo vision and sheet of light techniques. The main advantage of this method is that it overcomes the limitations of traditional methods and works robustly in ambient lighting condition. In this paper, we described a number of experiments carried out on different shapes and sizes of re?ective surface objects using the proposed method. The qualitative and quantitative reconstructed 3D pro?les of different shapes of re?ective surface objects are shown. In the end, the results are compared with the ground truth three-dimensional model of the acquited objects for accuracy evaluation.
关键词： Inspection,Re?ective Surface,Laser Line Projection,3D Measurements,Stereo Vision
更新于2025-09-19 17:13:59
-
Automatic Detection and Segmentation of Laser Stripes for Industrial Measurement
摘要： Laser stripe plays an important role in industrial vision measurement as the major auxiliary feature. Existing researches mainly focus on the application of small size parts. However, with the increase of field of view, it is difficult to extract laser stripes robustly in varying field measurement situations for the complex background, low proportion and uneven characteristic of laser stripes. To increase the measurement adaptability in complex environment, an automatic laser stripe detection and segmentation algorithm is proposed. First, the dataset is constructed by a large number of image patches collected in the field and laboratory, and laser stripe patches in the imbalanced dataset are expanded by data augmentation method. Next, the detection of the laser stripe is initially realized based on the training results of the convolutional neural network (CNN), and then the laser stripe is accurately detected by non-feature filtering criteria based on area constraints. Finally, a sub-regional feature clustering method is proposed to realize effective segmentation of uneven laser stripes. A large number of verification experiments have been carried out in both laboratory and field, and the results show that the proposed method can achieve automatic and accurate extraction of laser strips, which has strong adaptability to both the complex background in the field and the uneven brightness characteristic of laser stripes, satisfying the engineering requirements of large-scale parts field measurement.
关键词： detection and segmentation,stereo-vision measurement,CNN,laser stripe,large industrial part
更新于2025-09-16 10:30:52
-
Depth Information Enhancement using Block Matching and Image Pyramiding Stereo Vision Enabled RGB-D Sensor
摘要： Depth sensing devices enabled with an RGB camera, can be used to augment conventional images with depth information on a per-pixel basis. Currently available RGB-D sensors include the Asus Xtion Pro, Microsoft Kinect and Intel RealSenseTM. However, these sensors have certain limitations. Objects that are shiny, transparent or have an absorbing matte surface, create problems due to reflection. Also, there can be an interference in the IR pattern due to the use of multiple RGB-D cameras and the depth information is correctly interpreted only for short distances between the camera and the object. The proposed system, block matching stereo vision (BMSV) uses an RGB-D camera with rectified/non-rectified block matching and image pyramiding along with dynamic programming for human tracking and capture of accurate depth information from shiny/transparent objects. Here, the IR emitter generates a known IR pattern and the depth information is recovered by comparing the multiple views of the focused object. The depth map of the BMSV RGB-D camera and the resultant disparity map are fused. This fills any void regions that may have emerged due to interference or because of the reflective transparent surfaces and an enhanced dense stereo image is obtained. The proposed method is applied to a 3D realistic head model, a functional magnetic resonance image (fMRI) and the results are presented. Results showed an improvement in speed and accuracy of RGB-D sensors which in turn provided accurate depth information density irrespective of the object surface.
关键词： disparity estimation,Block matching,stereo vision,image pyramiding,RGB-D sensor,depth sensing
更新于2025-09-16 10:30:52
-
Stereo vision–based non-contact dynamic measurement method for vehicle braking time sequence
摘要： The braking stability of semi-trailer train is directly related to the safety of driving. The braking time sequence of vehicle in braking process directly determines the braking stability of the vehicle. To overcome the limitations of the existing technology, this paper proposes a novel non-contact vehicle braking time sequence measurement method. The proposed method is based on binocular stereo vision that has high precision, high efficiency and low cost. First, the binocular stereo vision measurement principle is introduced, the camera is calibrated by three-dimensional target and the Levenberg–Marquardt algorithm is used to optimize the calibration parameters. Then, a series of image processing techniques are used to extract the wheel marker and perform three-dimensional coordinate reconstruction. Finally, the universality and the applicability of the proposed method are verified by conducting repetitive experiments and accuracy experiments for the same vehicle and different types of vehicles according to the designed experimental process and the established evaluation index. The experimental results show that the method proposed in this paper can achieve fast detection, with high accuracy and good repeatability. Moreover, the method can be extended to various types of vehicles.
关键词： Levenberg-Marquardt algorithm,Stereo vision,braking time sequence,non-contact,dynamic detection
更新于2025-09-11 14:15:04