- 标题
- 摘要
- 关键词
- 实验方案
- 产品
过滤筛选
- 2018
- railway track
- kinematic characteristics
- stereo vision
- image blur
- distance estimation
- control system
- depth of field
- video monitoring
- hollows classification
- RGB-D camera
- Optoelectronic Information Science and Engineering
- Chongqing University of Posts and Telecommunications
- Bauman Moscow State Technical University
- Moscow State University of Railway Engineering
-
Color and depth image registration algorithm based on multi-vector-fields constraints
摘要： Image registration, which aim to establish a reliable feature relationship between images, is a critical problem in the field of image processing. In order to enhance the accuracy of color and depth image registration, this paper proposes an novel image registration algorithm based on multi-vector-fields constraints. We first initialize the edge information features of color and depth images, and establish putative correspondences based on edge information. Consider the correlation between the images, establish the functional relationships of the multi-vector-fields constraints based on the relationships. In the reproducing nuclear Hilbert space (RKHS), this constraint is added to the probability model, and the model parameters are optimized using the EM algorithm. Finally, the probability of corresponding edge points of the image is obtained. In order to further improve registration accuracy, this paper will change the input from one pair to two pairs and let the feature transformation relationship between images be iteratively evaluated using the parameter model. Taking publicly available RGB-D images as experimental subjects, results show that for single object image registration, the algorithm image registration accuracy in this paper is improved by about 5% compared with SC, ICP, and CPD algorithms. In addition, artificial noise was used to test the proposed algorithm’s anti-noise ability, results show that the proposed algorithm has superior anti-noise ability relative to SC, ICP and CPD algorithms.
关键词： Multi-vector-fields constraints,Image registration,EM algorithm,Depth image
更新于2025-09-23 15:23:52
-
Enhanced depth estimation of integral imaging using pixel blink rate
摘要： In this paper, we propose a new depth estimation technique with enhanced depth resolution in three-dimensional (3D) integral imaging. Typical integral imaging using a lenslet array can obtain elemental images with different perspectives by a single shot. However, the lateral and depth resolutions of the reconstructed 3D image may be low, which limit applications of integral imaging such as for object tracking, occlusion removal, and depth estimation. Especially, enhanced depth estimation may be required since depth is the most important information for all 3D applications. In this paper, we use the pixel rearrangement technique for visual quality enhancement of the reconstructed 3D image, and use pixel blink rate to evaluate the defocus area for depth estimation improvement. Our experimental results support that our method can enhance the depth resolution in typical integral imaging using lenslet arrays.
关键词： Depth estimation,Three-dimensional image acquisition,Three-dimensional image processing
更新于2025-09-23 15:23:52
-
Predicting postoperative delirium and postoperative cognitive decline with combined intraoperative electroencephalogram monitoring and cerebral near-infrared spectroscopy in patients undergoing cardiac interventions
摘要： Studies have associated electroencephalogram (EEG) suppression with postoperative delirium (POD) and postoperative cognitive decline (POCD). Otherwise, improving cerebral tissue oxygen saturation (rScO2) seems beneficial. No study has evaluated the impact of EEG suppression and decreased rScO2 on the incidence of POD and POCD when the intraoperative management of patients is performed with a depth-of-anesthesia (DOA) monitor and a cerebral oximetry. In this prospective study patients undergoing cardiac interventions were monitored with the NeuroSENSE? DOA monitor and bilateral cerebral oximetry. An algorithm was used to optimize cerebral oxygenation. EEG suppression was presented as total area under the curve (AUC) of suppression ratio (SR) > 0 s (AUC EEGSR>0s). Cerebral desaturation was defined as AUC of 25% drop of oximetry values as compared to baseline. POD was evaluated by the chart review method. POCD was defined as a Z-score ≤ 2 based on Mini Mental State Examination at baseline and day 5 or if the patient reported any cognitive decline at 3 and at 6 months postoperatively. Among the 1616 patients, 1513 underwent normothermic surgery and were further analyzed. POD and POCD were respectively evaluated in 1504 and 1350 patients of whom 303 (20%) and 270 (20%) were respectively diagnosed positive. Having experienced high magnitudes of EEG suppression (fourth quartile of AUC EEGSR>0s) was significantly associated with POD (OR = 2.247; 95% CI = 1.414–3.571; P = 0.001). Low rScO2 at the end of surgery was statistically associated with POCD (OR = 0.981; 95% CI = 0.965–0.997; P = 0.018). The results of our study show that the degree of intraoperative EEG suppression on one hand, and low rScO2 at the end of procedure on the other hand, are associated with respectively POD and POCD in patients undergoing cardiac interventions.
关键词： Burst suppression,Postoperative cognitive decline,Postoperative delirium,Cerebral oximetry,Depth-of-anesthesia monitor
更新于2025-09-23 15:23:52
-
Investigating the Aerosol Optical Depth and Angstrom Exponent and Their Relationships with Meteorological Parameters Over Lahore in Pakistan
摘要： In the present work, AERONET (AErosol RObotic NETwork) data of 2006–2014 have been used to analyze the variations in aerosol optical depth (AOD) at 500 nm and Angstrom exponent (440/870) (AE). In order to have an in-depth knowledge of aerosol variability, we have analyzed the association of aerosol properties with the meteorological parameters such as temperature, mean sea level pressure, rainfall, dew point, and dust storm frequency. Long-term observations of MODIS-AOD are also validated with AERONET-AOD over Lahore. The peak monthly mean value of AOD is found in July (1.00 ± 0.34) with the corresponding AE value of 0.85 ± 0.29 pointing toward the fact that desert/soil dust aerosols dominated the atmosphere of Lahore. The lowest value of AOD is found in February (0.47 ± 0.26) with the corresponding AE value of 1.22 ± 0.29 representing the presence of urban/industrial aerosols in the atmosphere over Lahore. The monthly mean AE value is found to be maximum in January (1.36 ± 0.15), whereas lowest value of AE is found in June (0.55 ± 0.25). AOD shows positive correlations with temperature, dew point, relative humidity, visibility, rain and dust storm frequency, and negative with mean sea level pressure and wind speed. AE exhibits positive correlations with relative humidity and mean sea level pressure, while with temperature, dew point, visibility, rain and dust storm frequency, it shows negative correlations.
关键词： Aerosol optical depth,Angstrom exponent,Meteorological parameters,Lahore
更新于2025-09-23 15:23:52
-
Nonrigid reconstruction of 3D breast surfaces with a low-cost RGBD camera for surgical planning and aesthetic evaluation
摘要： Accounting for 26% of all new cancer cases worldwide, breast cancer remains the most common form of cancer in women. Although early breast cancer has a favourable long-term prognosis, roughly a third of patients suffer from a suboptimal aesthetic outcome despite breast conserving cancer treatment. Clinical-quality 3D modelling of the breast surface therefore assumes an increasingly important role in advancing treatment planning, prediction and evaluation of breast cosmesis. Yet, existing 3D torso scanners are expensive and either infrastructure-heavy or subject to motion artefacts. In this paper we employ a single consumer-grade RGBD camera with an ICP-based registration approach to jointly align all points from a sequence of depth images non-rigidly. Subtle body deformation due to postural sway and respiration is successfully mitigated leading to a higher geometric accuracy through regularised locally affine transformations. We present results from 6 clinical cases where our method compares well with the gold standard and outperforms a previous approach. We show that our method produces better reconstructions qualitatively by visual assessment and quantitatively by consistently obtaining lower landmark error scores and yielding more accurate breast volume estimates.
关键词： Breast cancer treatment,Nonrigid registration,Depth camera,3D surface reconstruction,Aesthetic evaluation
更新于2025-09-23 15:23:52
-
[IEEE 2018 International Symposium on Electronics and Telecommunications (ISETC) - Timisoara, Romania (2018.11.8-2018.11.9)] 2018 International Symposium on Electronics and Telecommunications (ISETC) - Generating a Totally Focused Colour Image from Recordings of a Plenoptic Camera
摘要： This paper presents a method to calculate a totally focussed colour image from recordings of a plenoptic camera. The three colour channels have to be extracted from the raw image. The virtual depth is estimated analyzing the disparities of image points in different microlens images which correspond to the same object point. With the such created depth map, the image points are chosen from the microlens image with correct focus to create the totally focused colour image.
关键词： colour image,focused plenoptic camera,depth estimation
更新于2025-09-23 15:23:52
-
The effect of cut depth and distribution for abrasives on wafer surface morphology in diamond wire sawing of PV polycrystalline silicon
摘要： Due to the existence of an acid etch resistant thin amorphous silicon layer over the smooth grooves of the diamond wire sawing polycrystalline silicon wafer surface, the anti-re?ection e?ect is usually not ideal using the mature acidic texturization. The amorphous silicon layer will be produced on the machined surface by material ductile removal. Therefore, during the process of cutting photovoltaic polycrystalline silicon wafers, the material removed in the brittle way is expected and the surface topography of the wafers formed with the brittle fracture is better for the texture fabricating. In this paper, a mathematical model considering the in?uences of process parameters and wire saw parameters was developed based on indentation fracture mechanics. The variations of cutting groove pro?le formed by di?erent material removal modes were also included. The e?ect of abrasives distributed on the wire saw on material removal and surface formation of polysilicon was analyzed. The results showed that most of abrasives removed material with ductile removal mode, however, the volume of the material removed by abrasive in ductile mode is less than 10% of the total removal volume. Brittle fracture removal mode was still the major way of material removal in diamond wire sawing. With the same ratio of the feed rate and wire speed, the faster feed rate and wire speed will not only improve the cutting e?ciency, but also is easier to obtain a brittle fracture surface. There is a critical angle θc for the distribution of abrasives on the wire saw surface. Only when the position angle of the abrasive removing material in brittle mode is less than θc, the brittle fracture can be formed on the wafers surface.
关键词： Diamond wire sawing,Depth of cut,Material removal mode,Photovoltaic polycrystalline silicon
更新于2025-09-23 15:23:52
-
[IEEE 2018 24th International Conference on Pattern Recognition (ICPR) - Beijing, China (2018.8.20-2018.8.24)] 2018 24th International Conference on Pattern Recognition (ICPR) - Scale and Orientation Aware EPI-Patch Learning for Light Field Depth Estimation
摘要： Epipolar Plane Image (EPI) implies some important depth cues for light field depth estimation. Intuitively, the EPI patches with different spatial scales and orientations may exhibit different features and result in different estimation precision. In this paper, we discuss this issue and present a scale and orientation aware EPI-Patch learning model for depth estimation. We take the multi-orientation EPI patches of each pixel as input, and design two types of network structures for adaptive scale selection and orientation fusion. One type is a scale-aware structure, which feeds one orientation patch into a multi-layer feed-forward network with long and short skip connections. The other type is a shared-weight network for fusing the multi-orientation features. We demonstrate the effectiveness of our model by experiments on 4D Light Field Benchmark.
关键词： scale-aware,orientation-aware,depth estimation,deep learning,EPI,light field
更新于2025-09-23 15:23:52
-
[IEEE 2018 24th International Conference on Pattern Recognition (ICPR) - Beijing, China (2018.8.20-2018.8.24)] 2018 24th International Conference on Pattern Recognition (ICPR) - Accurate 3-D Reconstruction with RGB-D Cameras using Depth Map Fusion and Pose Refinement
摘要： Depth map fusion is an essential part in both stereo and RGB-D based 3-D reconstruction pipelines. Whether produced with a passive stereo reconstruction or using an active depth sensor, such as Microsoft Kinect, the depth maps have noise and may have poor initial registration. In this paper, we introduce a method which is capable of handling outliers, and especially, even significant registration errors. The proposed method first fuses a sequence of depth maps into a single non-redundant point cloud so that the redundant points are merged together by giving more weight to more certain measurements. Then, the original depth maps are re-registered to the fused point cloud to refine the original camera extrinsic parameters. The fusion is then performed again with the refined extrinsic parameters. This procedure is repeated until the result is satisfying or no significant changes happen between iterations. The method is robust to outliers and erroneous depth measurements as well as even significant depth map registration errors due to inaccurate initial camera poses.
关键词： point cloud,3-D reconstruction,RGB-D cameras,pose refinement,depth map fusion,registration errors
更新于2025-09-23 15:23:52
-
Estimation of spatiotemporal PM1.0 distributions in China by combining PM2.5 observations with satellite aerosol optical depth
摘要： Particulates smaller than 1.0 μm (PM1.0) have strong associations with public health and environment, and considerable exposure data should be obtained to understand the actual environmental burden. This study presented a PM1.0 estimation strategy based on the generalised regression neural network model. The proposed strategy combined ground-based observations of PM2.5 and satellite-derived aerosol optical depth (AOD) to estimate PM1.0 concentrations in China from July 2015 to June 2017. Results indicated that the PM1.0 estimates agreed well with the ground-based measurements with an R2 of 0.74, root mean square error of 19.0 μg/m3 and mean absolute error of 11.4 μg/m3 as calculated with the tenfold cross-validation method. The diurnal estimation performance displayed remarkable single-peak variation with the highest R2 of 0.80 at noon, and the seasonal estimation performance showed that the proposed method could effectively capture high-pollution events of PM1.0 in winter. Spatially, the most polluted areas were clustered in the North China Plain, where the average estimates presented a bimodal distribution during daytime. In addition, the quality of satellite-derived AOD, the robustness of the interpolation algorithm and the proportion of PM1.0 in PM2.5 were confirmed to affect the estimation accuracy of the proposed model.
关键词： Himawari-8,PM1.0,Neural network,Air pollution,Aerosol optical depth
更新于2025-09-23 15:23:52