Medical images application of contour extraction based on Hermite splines contour model

Active Contour Model or Snake model is an efficient method by which the users can extract the object contour of Region Of Interest （ROI）. In this paper, we present an improved method combining Hermite splines curve and Snake model that can be used as a tool for fast and intuitive contour extraction. We choose Hermite splines curve as a basic function of Snake contour curve and present its energy function. The optimization of energy minimization is performed hy Dynamic Programming technique. The validation results are presented, comparing the traditional Snake model and the HSCM, showing the similar performance of the latter. We can find that HSCM can overcome the non-convex constraints efficiently. Several medical images applications illustrate that Hermite Splines Contour Model （HSCM） is more efficient than traditional Snake model.

Contour Extraction of Skin Tumors Using Visual Attention and GVF-Snake Model

Contour extraction of skin tumors accurately is an important task for further feature generation of their borders and sur-faces to early diagnose melanomas. An integrated approach, combining visual attention model and GVF-snake, is pro-posed in the paper to provide a general framework for locating tumor boundaries in case of noise and boundaries with large concavity. For any skin image, the visual attention model is implemented to locate the Region of Interests (ROIs) based on saliency maps. Then an algorithm called GVF-snake is utilized to iteratively drive an initial contour, deriving from the extracted ROIs, towards real boundary of skin tumors by minimizing an energy function. It is shown from ex-periments that the proposed approach exceeds in two aspects compared with other contour-deforming methods: 1) ini-tial contours generated from saliency maps are definitely located at neighboring regions of real boundaries of skin tu-mors to speed up converges of contour deformation and achieve higher accuracy;2) the method is not sensitive to nois-es on skins and initial contours extracted.

CE-EEN-B0:Contour Extraction Based Extended EfficientNet-B0 for Brain Tumor Classification Using MRI Images

A brain tumor is the uncharacteristic progression of tissues in the brain.These are very deadly,and if it is not diagnosed at an early stage,it might shorten the affected patient’s life span.Hence,their classification and detection play a critical role in treatment.Traditional Brain tumor detection is done by biopsy which is quite challenging.It is usually not preferred at an early stage of the disease.The detection involvesMagneticResonance Imaging(MRI),which is essential for evaluating the tumor.This paper aims to identify and detect brain tumors based on their location in the brain.In order to achieve this,the paper proposes a model that uses an extended deep Convolutional Neural Network(CNN)named Contour Extraction based Extended EfficientNet-B0(CE-EEN-B0)which is a feed-forward neural network with the efficient net layers;three convolutional layers and max-pooling layers;and finally,the global average pooling layer.The site of tumors in the brain is one feature that determines its effect on the functioning of an individual.Thus,this CNN architecture classifies brain tumors into four categories:No tumor,Pituitary tumor,Meningioma tumor,andGlioma tumor.This network provides an accuracy of 97.24%,a precision of 96.65%,and an F1 score of 96.86%which is better than already existing pre-trained networks and aims to help health professionals to cross-diagnose an MRI image.This model will undoubtedly reduce the complications in detection and aid radiologists without taking invasive steps.

Extraction of coastline in high-resolution remote sensing images based on the active contour model

While executing tasks such as ocean pollution monitoring,maritime rescue,geographic mapping,and automatic navigation utilizing remote sensing images,the coastline feature should be determined.Traditional methods are not satisfactory to extract coastline in high-resolution panchromatic remote sensing image.Active contour model,also called snakes,have proven useful for interactive specification of image contours,so it is used as an effective coastlines extraction technique.Firstly,coastlines are detected by water segmentation and boundary tracking,which are considered initial contours to be optimized through active contour model.As better energy functions are developed,the power assist of snakes becomes effective.New internal energy has been done to reduce problems caused by convergence to local minima,and new external energy can greatly enlarge the capture region around features of interest.After normalization processing,energies are iterated using greedy algorithm to accelerate convergence rate.The experimental results encompassed examples in images and demonstrated the capabilities and efficiencies of the improvement.

Fast and Robust DCNN Based Lithography SEM Image Contour Extraction Models

Scanning electron microscope(SEM)metrology is critical in semiconductor manufacturing for patterning process quality assessment and monitoring.Besides feature width and feature-feature space dimension measurements from critical dimension SEM(CDSEM)images,visual inspection of SEM image also offers rich information on the quality of patterning.However,visual inspection alone leaves considerable room of ambiguity regarding patterning quality.To narrow the room of ambiguity and to obtain more statistically quantitative information on patterning quality,SEM-image contours are often extracted to serve such purposes.From contours,important information such as critical dimension and resist sidewall angle at any location can be estimated.Those geometrical information can be used for optical proximity correction(OPC)model verification and lithography hotspot detection,etc.Classical contour extraction algorithms based on local information have insufficient capability in dealing with noisy and low contrast images.To achieve reliable contours from noisy and low contrast images,information beyond local should be made use of as much as possible.In this regard,deep convolutional neural network(DCNN)has proven its great capability,as manifested in various computer vision tasks.Taking the full advantages of this maturing technology,we have designed a DCNN network and applied it to the task of extracting contours from noisy and low contrast SEM images.It turns out that the model is capable of separating the resist top and bottom contours reliably.In addition,the model does not generate false contours,it also can suppress the generation of broken contours when ambiguous area for contour extraction is small and non-detrimental.With advanced image alignment algorithm with sub-pixel accuracy,contours from different exposure fields of same process condition can be superposed to estimate process variation band,furthermore,stochastic effect induced edge placement variation statistics can easily be inferred from the extracted contours.

Edge Contour Extraction in MR Image Using Edgeflow Contour Model

This paper presents a new model for edge extraction of MR images, based on curve evolution and edgeflow techniques. At first the model for curve evolution is constructed, which automatically detect boundaries, and change of topology in terms of the edgeflow fields, and then the numerical approximation of the model is introduced, which is based on semi-implicit scheme to speed up the proposed approach. Finally, the numerical implementation is present and the experimental results show that the proposed model successfully extracts the edge contours, regardless of the heavy noise.

Image-based Extraction of Characteristic Value of Pathological Leaf Surface

[ Objective] Computer image processing technology was used to distinguish the angular leaf spot and spotted disease in the agricultural production. [Method] The computer vision technology was used to carry out chromatic research on the plant pathological characteristics. The color and texture were taken as the plant disease image characteristic parameter to extract the perimeter, area and the shape of the lesion image, thus carrying out the classification judgment on the disease image. [ Result] C IE1976H IS chorma percentage histogram method was adopted to extract chromaticity characteristic parameters, the process was simple and effective with fast operation speed, eliminating the effect of leaf size and shape. The statistical characteristic parameter of chorma histogram was analyzed to obtain chroma skewness, which could significantly distinguish different symptoms of disease. [ Conclusion] The study suggested that chroma skewness could be adopted as the characteristic parameter to distinguish spotted disease with angular leaf spot.

基于机器视觉的生态林场园林植物种植密度监控研究

科学合理的种植密度有助于维持生态林场的生物多样性,使土地利用率和植物产出效率得到显著提升,对园林经济效益和生态环境效益产生重要影响。基于此,通过遥感技术采集生态林场园林植物信息,利用机器视觉技术将原始生态林场园林帧图像转换为灰度图像,准确提取出生态林场园林植物轮廓;再引入决策树算法对园林植物进行分类并计算种植密度,结合不同植物种类的生长特性、园林设计要求以及生态功能需求,完成对植物密度的分析预警。结合实际林业试验环境对植物种植密度的设计、管理以及调控措施提出了建议,促进了生态林场的可持续发展。

基于Contourlet变换和PCNN的CT图像椎体解剖轮廓特征提取方法的研究

提出一种新的基于Contourlet变换和脉冲耦合神经网络(PCNN)的医学图像解剖轮廓特征提取算法。首先对原始椎体CT图像进行Contourlet变换,得到能稀疏表示图像边缘以及方向信息的子带和低频子带;然后结合PCNN对低频子带进行边缘轮廓细节提取,最后利用处理后的所有子带系数,通过Contourlet逆变换,提取出图像的边缘轮廓。实验将本算法提取的结果与Canny算子、区域生长法以及结合小波变换和PCNN的算法提取的图像边缘轮廓进行比较,结果表明新算法能够有效的实现医学图像解剖结构轮廓特征的提取。

A Dense Feature Iterative Fusion Network for Extracting Building Contours from Remote Sensing Imagery

Extracting building contours from aerial images is a fundamental task in remote sensing.Current building extraction methods cannot accurately extract building contour information and have errors in extracting small-scale buildings.This paper introduces a novel dense feature iterative(DFI)fusion network,denoted as DFINet,for extracting building contours.The network uses a DFI decoder to fuse semantic information at different scales and learns the building contour knowledge,producing the last features through iterative fusion.The dense feature fusion(DFF)module combines features at multiple scales.We employ the contour reconstruction(CR)module to access the final predictions.Extensive experiments validate the effectiveness of the DFINet on two different remote sensing datasets,INRIA aerial image dataset and Wuhan University(WHU)building dataset.On the INRIA aerial image dataset,our method achieves the highest intersection over union(IoU),overall accuracy(OA)and F 1 scores compared to other state-of-the-art methods.

Contour reconstruction of three-dimensional spiral CT damage image

In order to improve the diagnosis and analysis ability of 3D spiral CT and to reconstruct the contour of 3D spiral CT damage image,a contour reconstruction method based on sharpening template enhancement for 3D spiral CT damage image is proposed.This method uses the active contour LasSO model to extract the contour feature of the 3D spiral CT damage image and enhances the information by sharpening the template en.hancement technique and makes the noise separation of the 3D spiral CT damage image.The spiral CT image was procesed with ENT,and the statistical shape model of 3D spiral CT damage image was established.The.gradient algorithm is used to decompose the feature to realize the analysis and reconstruction of the contour feature of the 3D spiral CT damage image,so as to improve the adaptive feature matching ability and the ability to locate the abnormal feature points.The simulation results show that in the 3D spiral CT damage image contour reconstruction,the proposed method performs well in the feature matching of the output pixels,shortens the contour reconstruction time by 20/ms,and provides a strong ability to express the image information.The normalized reconstruction error of CES is 30%,which improves the recognition ability of 3D spiral CT damage image,and increases the signal-to noise ratio of peak output by 40 dB over other methods.

Automatic Body Feature Extraction from Front and Side Images

Human body feature extraction based on 2D images provides an efficient method for many applications, e.g. non-contact body size measurements, constructing 3D human model and recognizing human actions. In this paper a systematic approach is proposed to detect feature points of human body automatically from its front and side images. Firstly, an efficient approach for silhouette and contour detection is used to represent the contour curves of a human body shape with Freeman’s 8-connected chain codes. The contour curves are considered as a number of segments connected together. Then, a series of feature points on human body are extracted based on the specified rules by measuring the differences between the directions of the segments. In total, 101 feature points with clearly geometric properties (that rather accurately reflect the bump or turning of the contours) are extracted automatically, including 27 points corresponding to the definitions of the landmarks about garment measurements. Finally, the proposed approach was tested on ten human subjects and the entire 101 feature points with specific geography geometrical characteristics were correctly extracted, indicating an effective and robust performance.

五台山壁画人物线描图生成算法

五台山壁画图像因褪化而导致颜色不均匀、部分边缘轮廓像素灰度缺失,用传统的梯度法生成的线描图存在噪点多且线条不连续的问题。提出一种基于色调补偿及双卷积技术的壁画人物线描图生成算法。在计算梯度幅值时引入色调梯度进行有效的色调补偿,同时通过两种针对离散噪点的卷积模板实现去麻点功能。对五台山壁画人物图像的线描图生成实验的结果表明,该方法在去噪的基础上有效改善了轮廓线不连续的问题。

双目视觉的货车车厢完整尺寸测量

针对货车车厢完整尺寸的快速测量,提出了一种基于双目视觉的测量方法。依据相机成像模型和标定原理对双目相机进行标定实验,使用标定后的双目相机从货车侧后方采集货车车厢图像,再采用U 2-Net显著性目标检测算法对采集的货车车厢图像进行车厢轮廓提取,通过改进的基于图像分块的线段检测算法提取车厢边缘,结合车厢轮廓特征利用车厢边缘线段计算出特定的角点坐标,由三维坐标恢复后的车厢角点间距得到车厢尺寸。试验结果表明,车厢尺寸检测的平均误差小于4.06%,体积检测平均误差小于5.83%,可用于实际测量。

多源点云优化的城市三维模型构建

简易模型与传统复杂的单体模型相比,对存储、可视化、传输和后续应用技术要求低,敏感信息承载量少,且能满足大部分应用需求,具有更高的应用价值。考虑应用需求与技术现状,自然资源部将城市三维模型(L0D1.3级)纳入实景三维中国建设任务中。目前相关生产以半自动化采集为主,工程化应用存在技术流程烦琐、效率较低且对数据量限制较大等问题。本文立足山东实际,提出了一种基于多源点云优化的城市三维模型快速构建技术,在充分利用现有数据基础上有效提升了数据处理效率,大幅减少了生产成本,并在实景三维山东建设过程中得到了有效验证,具有较好的社会经济效益与推广价值。

考虑地域建筑特色偏好的城市主题公园景观规划设计

城市主题公园作为传承和促进城市文化发展的载体,其规划设计非常重要。基于当前城市主题公园景观规划设计中存在的不足,文章在考虑地域建筑特色偏好的基础上提出城市主题公园景观规划设计方法,并进行仿真实验。实验结果表明:该方法提取城市建筑屋顶轮廓较为精准,规划设计城市主题公园驳岸视线、指标以及防洪涝灾害效果均较好,具备较强的城市主题公园景观规划设计能力。

基于非完整点云法线滤波补偿的散货船舶舱口识别算法

自动装船系统是智能化港口建设的重要组成部分,能够大幅降低港口作业成本,提高经济效益。舱口识别作为自动装船任务的首要环节,成功率和识别精度是后续任务顺利进行的重要保障。由于港口激光雷达的数目和角度等问题,采集所得船舶点云数据时常出现缺失;此外船舶舱口附近经常有大量物料堆积,会使采集到的点云数据无法准确表达舱口的几何信息。由于上述港口实际装船作业中时常出现的问题,显著降低了现有算法的识别成功率,对自动装船作业造成了不良影响,因此迫切需要提升在船舶点云中存在物料干扰或舱口数据缺失的情况下的舱口识别成功率。基于船舶结构特征与自动装船过程中采集的点云数据分析,提出了基于非完整点云法线滤波补偿的散货船舶舱口识别算法。在使用港口实际采集点云所制作的数据集上进行了实验验证,识别成功率和识别精度较Miao和Li的舱口识别算法相比均有提升。实验结果表明,所提算法既能对舱口内物料噪声进行滤除,又能对数据缺失部分进行补偿,能够有效提升舱口识别效果。

基于级联U-Net的遥感影像道路分割和轮廓提取方法

针对基于深度学习的遥感图像道路信息提取模型往往只能输出单任务结果且多任务之间相关性利用不充分的问题,提出了一种基于级联U-Net的道路语义分割和轮廓联合检测方法,将道路语义分割后的特征图与原始图像融合后进行道路轮廓的提取,实现道路语义分割和边界轮廓的联合训练。首先使用U-Net网络结构提取光学遥感图像丰富的层次化特征,通过级联结构将特征串联融合,分别用于提取道路的语义类别和边界轮廓。其次在每级U-Net结构中引入注意力机制模块,进行空间上下文信息和深层次特征提取,改善网络提取过程中出现的细节模糊现象。最后,使用骰子系数和交叉熵误差组成的联合损失函数进行多任务整体训练,实现深度学习模型对遥感图像中道路语义类别和边界轮廓的同时提取。通过在加拿大渥太华城市地区的光学遥感数据集上进行实验,基于级联U-Net的道路信息联合提取方法在分割指标上分别获得了42%的精确度、58%的召回率、48.2%的F1分数以及71.6%的平均交并比,在道路检测指标上取得了0.896的全局最佳阈值(ODS)。结果表明,该模型在满足联合提取道路多任务信息的同时具有更优的检测精度。

一种采用改进Canny算子的机械臂识别与定位算法设计

为进一步提高机械臂图像处理的精确度,文中在改进Canny算子的基础上,提出一种适用于机械臂视觉伺服的图像识别与定位算法。该算法利用张正友标定法、灰度化、滤波及阈值分割等方法,完成了精确的相机标定和图像预处理,并在此基础上通过改进传统的Canny算子,降低了图像处理中边缘检测算法的检测误差,从而有效提高了机械臂的目标识别与定位精度。仿真实验结果表明,与传统算法相比,基于改进Canny算子的识别及定位算法具有更高的处理精度和更快的执行速度。

基于凹凸性和转向角的古陶瓷碎片二次匹配算法

碎片拼接是古陶瓷修复的关键工作,针对古陶瓷碎片形状随机、数量大、表面纹理弱且存在局部缺损而导致算法的精度较低、匹配时间较长等问题,提出一种基于凹凸性和转向角的古陶瓷碎片二次配算法。在提取古陶瓷碎片轮廓曲线的基础上,通过先后使用粗匹配和细匹配的二次匹配组合实现碎片的两两精确匹配。一次粗匹配先通过多边形逼近碎片轮廓曲线,以降低轮廓的复杂性,再提取多边形的顶点凹凸性和顶点转向角构建一次轮廓特征集合,最后利用凹凸互补性和遍历顶点对齐的双模态特征初次匹配算法来寻找大致匹配段,并得到粗匹配点集。二次细匹配先随机选取粗匹配点集中的任意相邻两点点对来提取碎片轮廓片段,以减少轮廓点数量并提高算法效率,再计算轮廓片段的轮廓转向角以提取二次轮廓特征集合,最后利用基于粒子群优化的二次匹配来搜索精确匹配段,并得到细匹配点集。实验结果表明,该算法对二维古陶瓷碎片的拼接效果较好,且具有较强的鲁棒性,拼接误差不超过2%,运行时间效率相比已有算法提高了8%~20%。