Learning hierarchical spatio-temporal pattern for human activity prediction

Human activity prediction has become increasingly valuable in many applications. This paper, initially from the perspective of cognition science, presents a novel approach to learning a hierarchical spatio-temporal pattern of human activities to predict ongoing activities from videos that contain only the onsets of the activities. Spatio-temporal pattern can be learned by a Hierarchical Self-Organizing Map (HSOM), which consists of two self-organizing maps (i.e., action map and actionlet map) connected via associative links trained by Hebbian learning. Ongoing activities can be predicted by Variable order Markov Model (VMM), which provides the means for capturing both large and small order Markov dependencies based on the training actionlet sequences. Experiments of the proposed method on four challenging 3D action datasets captured by commodity depth cameras show promising results.     

Unsupervised segmentation of hidden semi-Markov non-stationary chains

The Bayesian segmentation using Hidden Markov Chains (HMC) is widely used in various domains such as speech recognition, acoustics, biosciences, climatology, text recognition, automatic translation and image processing. On the one hand, hidden semi-Markov chains (HSMC), which extend HMC, have turned out to be of interest in many situations and have improved HMC-based results. On the other hand, the case of non-stationary data can pose an important problem in real-life situations, especially when the model parameters have to be estimated. The aim of this paper is to consider these two extensions simultaneously: we propose using a particular triplet Markov chain (TMC) to deal with non-stationary hidden semi-Markov chains. In addition, we consider a recent particular HSMC having the same computation complexity as the classical HMC. We propose a related parameter estimation method and the resulting unsupervised Bayesian segmentation is validated through experiments; in particular, a real radar image segmentations are provided.     

A multi-object image segmentation C–V model based on region division and gradient guide

The Chan–Vese (C–V) model is an ineffective method for processing images in which the intensity is inhomogeneous. This is especially true for multi-object segmentation, in which the target may be missed or excessively segmented. In addition, for images with rich texture information, the processing speed of the C–V is slow. To overcome these problems, this paper proposes an effective multi-object C–V segmentation model based on region division and gradient guide. First, a rapid initial contour search is conducted using Otsu’s method. This contour line becomes the initial contour for our multi-object segmentation C–V model based on a gradient guide. To achieve the multi-object segmentation the image is then converted to a single level set whose evolution is controlled using an adaptive gradient. The feasibility of the proposed model is analyzed theoretically, and a number of simulation experiments are conducted to validate its effectiveness.     

基于COLLATE融合多图谱的心脏电影MRI右心室分割

右心室分割对肺动脉高压等疾病的心功能分析具有重要的临床意义.然而，右心室心肌薄、易变且不规则，其传统的医学图像分割方法仍然未能取得突破性进展.本文提出基于COLLATE（Consensus Level，Labeler Accuracy and Truth Estimation）的多图谱分割方法，首先以归一化互信息为相似测度对目标图像和图谱集进行B样条配准以获取粗分割结果；然后利用COLLATE对粗分割结果进行融合；最后采用基于形状约束的区域生长算法修正出现错误的数据.10例临床心脏磁共振短轴电影图像被用于算法验证.本文还将使用基于COLLATE的多图谱分割方法得到的结果与深度学习算法及手动分割进行了比较.结果显示与深度学习算法比较，使用本文算法得到的射血分数（Ejection Fraction，EF）与手动分割更加一致和相关，表明该算法的分割结果有望辅助临床心脏功能诊断.     

基于最优阈值法的锅炉火焰图像处理的研究

工业锅炉炉膛内部是一个高温、高辐射、高灰度的环境，炉内燃料的燃烧过程伴随着剧烈而连续的发光发热的物理、化学反应。在锅炉火焰图像识别中，原始图像对该复杂环境变化非常敏感，并严重地影响图像的识别率。针对这一问题，本文介绍了一种提取锅炉火焰特征的最优阈值法。该方法依据统计学基本原理，对锅炉火焰自动识别的实际问题建立一个统计回归模型，并根据回归模型参数直接求取最优的分割阈值。     

A human visual system-driven image segmentation algorithm

This paper presents a novel image segmentation algorithm driven by human visual system (HVS) properties. Segmentation quality metrics, based on perceptual properties of HVS with respect to segmentation, are integrated into an energy function. The energy function encodes the HVS properties from both region-based and boundary-based perspectives, where the just-noticeable difference (JND) model is employed when calculating the difference between the image contents. Extensive experiments are carried out to compare the performances of three variations of the presented algorithm and several representative segmentation and clustering algorithms available in the literature. The results show superior performance of our approach.     

A high accuracy flow segmentation method in crowded scenes based on streakline

Flow segmentation based on similar motion patterns in crowded scenes remains an open problem in computer vision due to inherent complexity and vast diversity found in such scenes. To solve this problem, the streakline framework based on Lagrangian fluid dynamics had been proposed recently. However, this framework computed optical flow field using conventional optical flow method (Lucas Kanade method) which has poor anti-interference performance, and serious deviation would be brought to the computation of optical flow field. Moreover, our experimental results show that using the formulation of streak flow similarity in this framework can result in incorrect flow segmentation. Therefore, we combine this framework with a high accurate variational model, and modify the corresponding formulation of streak flow similarity after analyzing the streakline framework in detail. Finally, an improved method is proposed to solve flow segmentation in crowded scenes. Experiments are done to compare these two methods and results verify the validity and accuracy of our method.     

基于视觉信息的移动机器人控制系统设计

机器人的视觉相当于机器人的眼睛,机器人视觉系统是当下研究热点。本文主要研究基于视觉信息的机器人系统,通过对获得的图像信息的处理来实现对机器人的控制。详细讨论了摄像机标定、图像处理等方面的设计,实现了对机器人控制系统的设计。     

一种改进的基于Viterbi的语音切分算法

主要针对文本提示型说话人识别中语音切分高精确度要求的问题,在利用Viterbi算法的语音切分基础上,提出了向后平滑搜索多帧能量极小值的语音切分方法。该算法首先对0~9的每个数字建立模型,然后利用Viterbi算法对随机数字串进行切分得到初始切分点,最后利用搜索多帧能量极小值的方法更新原始切分点。实验表明,相比于传统的切分算法,在误差范围小于20 ms之内,改进算法的切分准确率由82.1%提高到88%。