Search

Face classification is important in many applications such as surveillance, border control, and security systems. However, wide variations in environments such as insufficient light, large distances or pose angles make the task challenging. Depth sensors are added with RGB cameras for improving classification accuracy but commercial RGB-D sensors are most targeted for indoors applications. In this paper, we present and design a Chaos LiDAR depth sesnor that provides high-precision depth images through intelligent correlation processing for both indoors and outdoors applications. Our Chaos LiDAR depth sensor detects range from 2 to 40 meters with precision around 8mm at 20-meter. With the Chaos LiDAR depth as input, we design a RGB-D based face classification embedded CNN (eCNN) model for wide range applications such as dim illumination, various distances and large poses. Our Chaos LiDAR increases around 14.27% classification accuracy compared to RealSense D435i for distance from 3 to 5 meter. The eCNN face classification subsystem is implemented in Xilinx ZCU 102 and achieves 11.11 ms inference time. The eCNN engine achieves a peak throughput at 614.4 GOPS. The overall system including Chaos LiDAR, correlation and eCNN FPGA achieves face classification inference rate of 10fps.

Increasing the vertical dimension of occlusion (VDO) is a common procedure in complicated prosthodontic treatment. The swallowing technique had been verified as a functional method to determine the VDO. The purpose of this study was to investigate the association between increasing VDO and mandibular movement during swallowing.

AbstractThe shape-from-silhouette (SFS) method has been widely used in 3D shape reconstruction. It uses silhouettes of a series of 2D images captured from multiple viewpoints of an object to generate a 3D model that describes the visual hull of the object. The SFS method faces an inherent problem that virtual features appear all over the model. In addition, concavities on the object may wrongly be modeled as convex shapes because they are invisible on image silhouettes. The purpose of this study is to propose a method to generate a 3D model from silhouettes of multiple images and propose a quality improvement method to overcome the above-mentioned problems. The 3D modeling method focuses on accurate evaluation of 3D points intersected by all polyhedra from different views and the removal of poor meshes on triangulation. The quality improvement method is essentially an iterative procedure, which for smoothing the model and eliminating virtual features and artifacts, while preserving the consistency of all silhouettes. The proposed method is to be used for product presentations in e-commerce, in which the 3D model must be covered with color texture of an object. Several examples are presented to illustrate the capability of the proposed method.

Piezoelectric energy harvesting is a promising technique for scavenging ambient mechanical motion for driving compact, low-power, multi-functional electronic devices. To adapt to various ambient surroundings, the geometric configurations and sizes varied in wide ranges with high operational reliabilities and piezoelectric performance have been regarded as a key for piezoelectric harvester design. Herein, by applying a normal force, we report an innovative structure for harvesting electric energy from bending the obliquely aligned GaN piezoelectric nanorods (NRs) that are integrated in the vertically integrated nanogenerator (VING). The single-crystalline GaN NRs used here were successfully synthesized with obliquely alignments on the pyramided Si substrate by plasma-assisted molecular beam epitaxy (PA-MBE). Using conductive atomic force microscope (c-AFM), a remarkable change in the Schottky barrier height (SBH) between the tip and GaN NR is observed upon bending an oblique-aligned GaN NR. This demonstrates that a remarkably enhanced piezoelectric performance of GaN NRs can be achieved by coupling a lateral force. We anticipate that this work will provide an efficient approach for coupling the lateral loading to enhance the electric potential in piezoelectric NRs-embedded VING, and thus open a new path for efficiently generating electric energy.

PROBLEM: Elevated serum inhibin levels have been found in ovarian cancer patients; however, the source of the elevated inhibin is uncertain. Previous studies of activin in human ovarian cancer suggest that activin may promote the growth of the tumor. The aims of this study were to examine the source of elevated inhibin from ovarian epithelial tumors (OETs) and to preliminarily investigate the role of the gonadotropin–inhibin/activin relationship in the development of OET.
 METHOD OF STUDY: The protein and mRNA expression of α and βA subunits of inhibin/activin were examined by immunohistochemistry (IHC) and reverse transcription polymerase chain reaction (RT‐PCR) in 120 OETs, including 30 benign cystadenomas, 30 borderline tumors, and 60 carcinomas. Stromal and epithelial cells were microdissected from 23 OETs to further examine the expression of α and βA subunits by RT‐PCR. Dimeric inhibin A and activin A production were measured by using the two‐site ELISA from three OET cell lines in culture under treatment of follicle‐stimulating hormone (FSH) and luteinizing hormone (LH).
 RESULTS: βA subunit was expressed in the epithelial component of 100% of the cystadenomas, in 80% of borderline tumors, and in 75% of the carcinomas, but not in tumor stroma. Inhibin α expression was not found in the epithelium of all OETs studied, but focal inhibin α immunoreactivity was seen in the tumor stroma (mainly luteinized stromal cells) in the majority of cases. Dimeric activin A was produced by all of the three OET cell lines with a 1.5–1.9‐fold increment after FSH stimulation. However, activin A production was not augmented by LH treatment. No inhibin A was produced by the three OET cell lines with or without gonadotropin stimulation.
 CONCLUSIONS: The stroma of OET is the major source in the production of inhibin α (monomer). Dimeric inhibin A production may be the result of combined efforts of the tumor stroma (α subunit) and epithelium (βA subunit). Cellular, compartmental expression of inhibin and activin subunits may play a role in the development of OET, although the mechanism remains undefined. The unopposed activin A production stimulated by FSH in OET cell lines suggests that activin production may represent one of the cellular mechanisms of growth promotion by FSH.

RNA polymerase III in an extract of HeLa cells transcribes RNA 5S in size from genomic bovine DNA template. This RNA represents a major fraction of the RNA synthesized. 5S RNA is not transcribed from bovine chromatin at equivalent concentrations and RNA the size of tRNA or tRNA precursor is not detected. At low UTP concentrations RNA slightly smaller in size than 5S is synthesized in addition to 5S RNA.

This study aimed to determine the effectiveness of using noninvasive arterial pulse-wave and laser-Doppler flowmetry (LDF) measurements to discriminate between colorectal-cancer (CC) patients and healthy control subjects. Radial-artery blood pressure waveform (BPW), finger photoplethysmography (PPG), and skin-surface LDF signals were measured noninvasively in 12 CC patients and 25 control subjects. Beat-to-beat, spectral, and variability analyses were applied to 20-minute-long recorded signals. Significant intergroup differences were found. In BPW, C1–C4amplitude indices were significantly larger while P2–P8phase-angle indices were significantly smaller in the CC patients than in the controls. The PPG and LDF variability indices were significantly larger and smaller, respectively, in CC patients. The relative energy contributions of the endothelial-, neural-, and myogenic-related frequency bands in LDF were significantly smaller in CC patients. The present findings indicate that pulse and LDF waveform analysis can be used to evaluate the arterial pulse-wave transmission condition, the responses of the blood-flow perfusion, and its regulatory activities in CC patients. There could be some similarities and differences in the present indices for different types of cancer. These findings could be utilized in the development of a rapid, noninvasive, and objective technique for evaluating the CC-induced blood-flow responses.