•Accurately predict TSV in offices in real time with physiological signals from wristband.•Develop and validate a smart HVAC control system in multi-occupant offices by using physiological and indoor ...environmental parameters.•Improve the overall thermal comfort in multi-occupant offices by using the wristband control system.•Energy use of the developed wristband control system was almost the same as using constant set point.
Since people spend most of their time indoors, it is important to create comfortable indoor environments for building occupants. However, unsuitable thermostat settings lead to energy waste and an undesirable indoor environment, especially in multi-occupant rooms. This study aimed to develop and validate a control strategy for the HVAC systems in multi-occupant offices using physiological parameters measured by wristbands. We used an ANN model to predict thermal sensation from indoor environmental and physiological parameters such as air temperature, relative humidity, clothing level, wrist skin temperature, skin relative humidity and heart rate. The model was trained by data collected in seven multi-occupant offices in the course of a year, and it was able to predict the thermal sensation with high accuracy. Next, we developed a control strategy for the HVAC system to improve the thermal comfort of all the occupants in the room. The control system was smart and could adjust the thermostat set point automatically in real time. We validated the system by means of both experiments and numerical simulations. In most cases, we improved the occupants’ thermal comfort level. After using the wristband control, over half of the occupants experienced a neutral sensation, and fewer than 5% still felt uncomfortable. The energy consumption by the HVAC system with the wristband control was almost the same as when the constant set point was used. After coupling with occupancy-based control by means of lighting sensors or wristband Bluetooth, the heating and cooling loads were reduced by 90% and 30%, respectively, in interior offices. Therefore, the smart HVAC control system can effectively control the indoor environment for thermal comfort and energy saving.
Detecting vehicles in aerial imagery plays an important role in a wide range of applications. The current vehicle detection methods are mostly based on sliding-window search and handcrafted or ...shallow-learning-based features, having limited description capability and heavy computational costs. Recently, due to the powerful feature representations, region convolutional neural networks (CNN) based detection methods have achieved state-of-the-art performance in computer vision, especially Faster R-CNN. However, directly using it for vehicle detection in aerial images has many limitations: (1) region proposal network (RPN) in Faster R-CNN has poor performance for accurately locating small-sized vehicles, due to the relatively coarse feature maps; and (2) the classifier after RPN cannot distinguish vehicles and complex backgrounds well. In this study, an improved detection method based on Faster R-CNN is proposed in order to accomplish the two challenges mentioned above. Firstly, to improve the recall, we employ a hyper region proposal network (HRPN) to extract vehicle-like targets with a combination of hierarchical feature maps. Then, we replace the classifier after RPN by a cascade of boosted classifiers to verify the candidate regions, aiming at reducing false detection by negative example mining. We evaluate our method on the Munich vehicle dataset and the collected vehicle dataset, with improvements in accuracy and robustness compared to existing methods.
Vehicle detection in aerial images, being an interesting but challenging problem, plays an important role for a wide range of applications. Traditional methods are based on sliding-window search and ...handcrafted or shallow-learning-based features with heavy computational costs and limited representation power. Recently, deep learning algorithms, especially region-based convolutional neural networks (R-CNNs), have achieved state-of-the-art detection performance in computer vision. However, several challenges limit the applications of R-CNNs in vehicle detection from aerial images: 1) vehicles in large-scale aerial images are relatively small in size, and R-CNNs have poor localization performance with small objects; 2) R-CNNs are particularly designed for detecting the bounding box of the targets without extracting attributes; 3) manual annotation is generally expensive and the available manual annotation of vehicles for training R-CNNs are not sufficient in number. To address these problems, this paper proposes a fast and accurate vehicle detection framework. On one hand, to accurately extract vehicle-like targets, we developed an accurate-vehicle-proposal-network (AVPN) based on hyper feature map which combines hierarchical feature maps that are more accurate for small object detection. On the other hand, we propose a coupled R-CNN method, which combines an AVPN and a vehicle attribute learning network to extract the vehicle's location and attributes simultaneously. For original large-scale aerial images with limited manual annotations, we use cropped image blocks for training with data augmentation to avoid overfitting. Comprehensive evaluations on the public Munich vehicle dataset and the collected vehicle dataset demonstrate the accuracy and effectiveness of the proposed method.
Single capacitive coupled wireless power transfer (SCC-WPT) technology only needs one pair of metal plates for energy transmission. This provides a unique advantage in two-dimensional plane dynamic ...wireless power transfer. However, in practical application, this technology has a clear challenge in improving the transmission power ratings and efficiency. This article studies the power transfer mechanism of the SCC-WPT system based on stray capacitance and self-capacitance. A universal system model has been investigated, where the equivalent circuit model of the system is proposed. The two-port network of the coupler critical factors that limit the transmission power of the system has been analyzed. A new LC-CL compensation topology is proposed to achieve a phase difference of voltage between the primary side and secondary side to 90°, which improves the transmission power of the SCC-WPT system. In addition, the resonant operating condition of the system together with the efficiency is analyzed. Finally, an experimental prototype is built under the guidance of the analysis. Both simulation and experimental results verify the validity of the proposed model and the topology. The experimental results show that the prototype achieves 266 W output power with 62% dc-dc efficiency.
Vehicle detection with orientation estimation in aerial images has received widespread interest as it is important for intelligent traffic management. This is a challenging task, not only because of ...the complex background and relatively small size of the target, but also the various orientations of vehicles in aerial images captured from the top view. The existing methods for oriented vehicle detection need several post-processing steps to generate final detection results with orientation, which are not efficient enough. Moreover, they can only get discrete orientation information for each target. In this paper, we present an end-to-end single convolutional neural network to generate arbitrarily-oriented detection results directly. Our approach, named Oriented_SSD (Single Shot MultiBox Detector, SSD), uses a set of default boxes with various scales on each feature map location to produce detection bounding boxes. Meanwhile, offsets are predicted for each default box to better match the object shape, which contain the angle parameter for oriented bounding boxes’ generation. Evaluation results on the public DLR Vehicle Aerial dataset and Vehicle Detection in Aerial Imagery (VEDAI) dataset demonstrate that our method can detect both the location and orientation of the vehicle with high accuracy and fast speed. For test images in the DLR Vehicle Aerial dataset with a size of 5616 × 3744 , our method achieves 76.1% average precision (AP) and 78.7% correct direction classification at 5.17 s on an NVIDIA GTX-1060.
The nanohydroxyapatite/polyamide 66 (n-HA/PA66) cage, a bioactive nonmetal cage, is fabricated in a hollow cylindrical shape and has been widely used for decades with good clinical outcomes for ...anterior cervical fusion. However, there remain some radiological complications, such as a slightly high subsidence rate. To improve the clinical outcomes, the improved n-HA/PA66 cage now has been developed into a trapezoidal and wedge shape, a better biomechanical shape matching the cervical spine that is similar to that of the PEEK cage. However, there have been no long-term comparisons of the improved n-HA/PA66 cage and PEEK cage in anterior cervical reconstruction.
Fifty-eight patients who underwent single-level anterior cervical decompression and fusion (ACDF) with the improved n-HA/PA66 cage (n-HA/PA66 group) were matched with patients with the PEEK cage (PEEK group) by clinical presentation, segment, age and sex. All patients underwent a minimum of 6 years of follow-up. The radiographic parameters (cage subsidence, fusion status, cervical lordosis, and segmental sagittal alignment) and clinical parameters (10-point visual analogue scale, Neck Disability Index and Japanese Orthopedic Association scores) from patients were evaluated before surgery, immediately after surgery, and at the latest follow-up.
The n-HA/PA66 and PEEK groups were well matched in terms of clinical presentation, segment, age, and sex at surgery. The n-HA/PA66 and PEEK cages had similar fusion rates at 6 months postoperatively (n-HA/PA66: 58.6% vs. PEEK: 51.7%, P = 0.455) and at the last follow-up (n-HA/PA66: 96.6% vs. PEEK: 93.1%, P = 0.402). The respective cage subsidence rates in the n-HA/PA66 and PEEK groups were 6.9 and 12.1% (P = 0.342). The correction of SA was similar between the groups at the final follow-up (n-HA/PA66: 4.29 ± 1.99 vs. PEEK: 3.99 ± 2.59 P = 0.464). There were no significant differences between the two groups in mean cervical lordosis, visual analogue scale scores of the neck and arm, NDI scores, JOA scores or patients' overall satisfaction at the final follow-up.
After single-level ACDF, the improved n-HA/PA66 cage had similar excellent results in both radiological and clinical outcomes compared with the PEEK cage over 6 years of follow-up. According to these results, the improved n-HA/PA66 cage and the PEEK cage could be comparable for ACDF.
This paper proposes a multi-line structured light stripe clustering method based on a custom iterative window to address the issue of stripes confusion caused by stripe breakage and noise points ...during the multi-line laser stripe matching process in binocular stereo vision.
Firstly, the paper segments the measured object from the carrier plane. Secondly, the extracted centerlines are clustered using the custom iterative window. Finally, the stripes are numbered one by one after clustering the left and right images using barycentric coordinates of the stripes, based on the property of sequential consistency constraint, to complete the stereo matching of the stripes with the same number.
The experimental results show that the method can effectively cluster the centerline stripes on the measured object and solve the problem of stripes mis-corresponding when stereo matching of multi-line structured light stripes.
Abstract
Background
In clinical practice, there are a significant percentage of Lenke 2 AIS patients suffered from medial shoulder imbalance (MSI) despite achieving good lateral shoulder balance ...(LSB) following surgery. However, there are few studies evaluating the features of the medial shoulder. The objective of this study was to determine the incidence and independent risk factors of MSI with LSB after Lenke 2 AIS corrective surgery.
Methods
One hundred and twenty Lenke 2 AIS patients with LSB at the last follow-up were reviewed from 2009 to 2018. Preoperative, and 3-month and the last postoperative follow-up radiographs were measured using a number of specific measurements. At the last follow-up, patients were divided into medial shoulder balance (MSB) group and the MSI group according to whether the T1 tilt was greater than 3°. A stepwise multiple linear regression analysis was used to examine the independent risk factors for MSI. Scoliosis Research Society (SRS)-30 questionnaire was used to assess clinical outcomes.
Results
Up to 69.2% of patients suffered from MSI with LSB after Lenke Type 2 AIS corrective surgery. Multiple regression showed that postoperative upper instrumented vertebra tilt (UIVt), proximal thoracic curve (PTC), the ratio of PTC and main thoracic curves (PTC/MTC) and T2 vertebra rotation ratio (T2-VR) were significant predictors for MSI (UIVt: b = 0.398, p < 0.001; PTC/MTC: b = 2.085, p < 0.001; PTC: b = 0.155, p < 0.001; T2-VR: b = 3.536, p = 0.008; adjusted R
2
= 0.711). 72 patients completed the SRS-30 questionnaire survey, and the MSB group were scored the higher (p ≤ 0.001) in self-image domain (4.18 ± 0.43 vs. 3.70 ± 0.35), satisfaction domain (4.39 ± 0.54 vs. 3.95 ± 0.46) and total average (4.31 ± 0.23 vs. 4.11 ± 0.19).
Conclusion
Although the patients with Lenke 2 AIS achieve LSB after corrective surgery, up to 69.2% of them suffered from MSI. Postoperative UIVt, PTC, PTC/MTC and T2-VR were significant predictors for MSI. Sufficient correction of these variables may facilitate the achievement of MSB.
Display omitted
•A heterogeneous structure was built via a Voronoi-Thiessen based parametric modeling method.•Heterogeneity had little influence on porosity, thus, mechanical/hydrodynamic ...properties.•Irregular local geometry mimicked cancellous bone structure and promoted cellular behavior.•50% heterogeneity (SR structure) kept a balance of structure integrity and osteogenesis.
In nature, many natural porous tissues of animals or plants follow heterogeneous structures pattern. Including turtle back pattern, leaf veins, wood porous fiber and porous cancellous bone are classic heterogeneous structures. In order to better design bionic microenvironment porous bone implants, herein, a bionic porous structure based on Voronoi-Thiessen diagram to create heterogeneous topology microenvironment promoting bone regeneration were proposed. Porous structures with tunable heterogeneity were built and achieved by selective laser melting (SLM). The topology analysis revealed the introduce of heterogeneity brought wide pore size distribution and irregular topology. The mechanical and hydrodynamics properties were predesigned and simulated by finite element analysis (FEA). The heterogeneity hardly influenced the overall strength, stiffness and permeability but the irregular stress and fluid shear rate distribution in FEA results indicated fluctuating local stiffness and permeability. In vitro and in vivo experimental results indicated that Quasi-homogeneous structure had irregular topology mimicking that of the natural bone structure and semi-heterogenous structure shown better osteogenesis. It is found that the scaffold heterogeneous microenvironment factors play important roles for bone regeneration, and which is crucial for the development of new orthopedic implants.
Display omitted
•Reinforcement learning of occupant behavior in adjusting thermostat and clothinglevel in office building.•Transfer learning model of occupant behavior cross-building without ...data.•Transfer learning among buildings of the same type was better than among different types .•The variation of energy use predicted by RL model was smaller than ANN model.
Occupant behavior plays an important role in the evaluation of building performance. However, many contextual factors, such as occupancy, mechanical system and interior design, have a significant impact on occupant behavior. Most previous studies have built data-driven behavior models, which have limited scalability and generalization capability. Our investigation built a policy-based reinforcement learning (RL) model for the behavior of adjusting the thermostat and clothing level. Occupant behavior was modelled as a Markov decision process (MDP). The action and state space in the MDP contained occupant behavior and various impact parameters. The goal of the occupant behavior was a more comfortable environment, and we modelled the reward for the adjustment action as the absolute difference in the thermal sensation vote (TSV) before and after the action. We used Q-learning to train the RL model in MATLAB and validated the model with collected data. After training, the model predicted the behavior of adjusting the thermostat set point with R2 from 0.75 to 0.8, and the mean absolute error (MAE) was less than 1.1 °C (2 °F) in an office building. This study also transferred the behavior knowledge of the RL model to other office buildings with different HVAC control systems. The transfer learning model predicted the occupant behavior with R2 from 0.73 to 0.8, and the MAE was less than 1.1 °C (2 °F) most of the time. Going from office buildings to residential buildings, the transfer learning model also had an R2 over 0.6. Therefore, the RL model combined with transfer learning was able to predict the building occupant behavior accurately with good scalability, and without the need for data collection.