A novel algorithm for automatic localization of human eyes

Based on geometrical facial features and image segmentation, we present a novel algorithm for automatic localization of human eyes in grayscale or color still images with complex background. Firstly, a determination criterion of eye location is established by the prior knowledge of geometrical facial features. Secondly, a range of threshold values that would separate eye blocks from others in a segmented face image (i.e., a binary image) are estimated. Thirdly, with the progressive increase of the threshold by an appropriate step in that range, once two eye blocks appear from the segmented image, they will be detected by the determination criterion of eye location. Finally, the 2D correlation coefficient is used as a symmetry similarity measure to check the factuality of the two detected eyes. To avoid the background interference, skin color segmentation can be applied in order to enhance the accuracy of eye detection. The experimental results demonstrate the high efficiency of the algorithm and correct localization rate.

A Reconfigurable Manufacturing System for In-site Machining Component with Large Scale

To cope with various unpredictable changes in large scale parts,the concept of reconfigurable manufacturing system (RMS) for machining these components is presented.Considering with large-size space measurement and the fixed-free manufacture mode,an automatically localizing machining method for large scale part is studied in this paper,and the architecture of the RMS for machining large scale parts is proposed.According to the method and structure,the automatically localizing model is established.The theoretical analysis and simulation examples demonstrate the feasibility and validity of the proposed method,and the results indicate that the method is suitable and effective for machining large scale components in significant scientific projects.

Detection of Left Ventricular Cavity from Cardiac MRI Images Using Faster R-CNN

The automatic localization of the left ventricle(LV)in short-axis magnetic resonance(MR)images is a required step to process cardiac images using convolutional neural networks for the extraction of a region of interest(ROI).The precise extraction of the LV’s ROI from cardiac MRI images is crucial for detecting heart disorders via cardiac segmentation or registration.Nevertheless,this task appears to be intricate due to the diversities in the size and shape of the LV and the scattering of surrounding tissues across different slices.Thus,this study proposed a region-based convolutional network(Faster R-CNN)for the LV localization from short-axis cardiac MRI images using a region proposal network(RPN)integrated with deep feature classification and regression.Themodel was trained using images with corresponding bounding boxes(labels)around the LV,and various experiments were applied to select the appropriate layers and set the suitable hyper-parameters.The experimental findings showthat the proposed modelwas adequate,with accuracy,precision,recall,and F1 score values of 0.91,0.94,0.95,and 0.95,respectively.This model also allows the cropping of the detected area of LV,which is vital in reducing the computational cost and time during segmentation and classification procedures.Therefore,itwould be an ideal model and clinically applicable for diagnosing cardiac diseases.

An Efficient Algorithm for Mobile Localization in Sensor Networks

Node localization is a fundamental problem in wireless sensor network.There are many existing algorithms to estimate the locations of the nodes.However,most of the methods did not consider the presence of obstacles.In practice,obstacles will lead to blockage and reflection of communication signals between sensor nodes.Therefore,the presence of obstacles will greatly affect the localization result.In this paper,we implement an obstacle-handling algorithm based on the localization tool developed by MIT,The experimental result shows that the enhanced algorithm can reduce the average distance error by up to 46 %,compared to the original algorithm.

A Protocol for Digitalized Collection of Traditional Chinese Medicine(TCM)Pulse Information Using Bionic Pulse Diagnosis Equipment

Pulse diagnosis equipment used in Traditional Chinese Medicine(TCM)has long been developed for collecting pulse information and in TCM research.However,it is still difficult to implement pulse taking automatically or efficiently in clinical practice.Here,we present a digital protocol for TCM pulse information collection based on bionic pulse diagnosis equipment,which ensures high efficiency,reliability and data integrity of pulse diagnosis information.A four-degree-of-freedom pulse taking platform together with a wrist bracket can satisfy the spatial positioning and angle requirements for individually adaptive pulse acquisition.Three-dimensional reconstruction of a wrist surface and an image localization model are combined to provide coordinates of the acquisition position and detection direction automatically.Three series elastic joints can not only simulate the TCM pulse taking method that“Three fingers in a straight line,the middle finger determining the‘Guan’location and finger pulp pressing on the radial artery,”but also simultaneously carry out the force-controlled multi-gradient pressing process.In terms of pulse information integrity,this proposed protocol can generate rich pulse information,including basic individual information,pulse localization distribution,multi-gradient dynamic pulse force time series,and objective pulse parameters,which can help establish the fundamental data sets that are required as the pulse phenotype for subsequent comprehensive analysis of pulse diagnosis.The implementation of this scheme is beneficial to promote the standardization of the digitalized collection of pulse information,the effectiveness of detecting abnormal health status,and the promotion of the fundamental and clinical research of TCM,such as TCM pulse phenomics.