| 1. | Product - line feature
产品线特色决策 |
| 2. | Global motion estimation using multi - line feature for image stabilization application
多直线全局运动估计及其在图像稳定中的应用 |
| 3. | A multi - threshold based morphological approach for extracting coastal line feature from remote sensed images
一种基于多阈值的形态学提取遥感图象海岸线特征方法 |
| 4. | Abstract : the paper presents a new approach to matching an image to a model based on straight line features
文摘:提出了一种新的基于直线特征的图像与模型匹配方法。 |
| 5. | Straight - line features are prominent in most man - made environments . they can be detected and tracked easily in image data
直线在大多数人造环境中是最基本的特征，直线可以很稳定地检测和跟踪。 |
| 6. | 12 zhang d , shu w . two novel characteristics in palmprintverification : datum point invariance and line feature matching . pattern recognition , 1999 , 32 : 691 - 702
在掌纹验证和辨识中，基于小波能量特征的方法分别达到了99 . 24 %和99 . 45 %的精度，这些结果优于其他的掌纹识别方法。 |
| 7. | In the end of this paper , we introduce that in the context of weather radar observations the enhancement of thin - line features in meteorological radar reflectivity images is addressed using a wavelet - based analysis . passing the thin line highlighted image into forecast system to complete the feature identification , we can get basis element for thunderstorm forecasting
本文最后介绍了采用小波变换的方法对滨州雷达资料的反射率图像进行处理，得到特征线增强后的图像，将其输入自动预报系统实现特征识别，可为雷暴预报等提供某些初步识别判据。 |
| 8. | Basing on background of urban transportation planning , the thesis presents a stdm of urban transportation and gives a special emphasis to spatiotemporal representation of point feature , line feature and area feature of transportation planning basic information , treatment of disaggregate travel data , and management of planning project
本论文以城市交通规划为背景，对城市交通规划时空数据模型的设计方法进行了讨论，提出了一个城市交通规划时空数据模型。论文着重对交通规划基础信息面状要素、线。 。 |
| 9. | By taking advantages of epipolar line features and depth discontinuities in reference 中国科学院 软件 研究所 博士 学位 论文 基于 图 象 的 快速 绘制 技术 的 研究 images ， an efficient inverse wmping algorithm is pfoposed in chapter 3 for gcnerating nagcs of novel views by combining multiple eference images 帆 enhm different vie 呷 oints because continuous segnents determi 。 d by pairs ofedge pixels at co 。 spending epipolar lines are order kept ， only pairs of edge pixels in the reference 渝 明 e e necess 叨 口 cowute to obtain generalized disparity of all points in the desired image as a result ， sighficant acceleraion could be made in the endering pfo 比 鸵 two accelerating techiq 此 s e presented in this algori 山 mb accelerate the hole illing process his algorithm extends the reference images rom projection of single col ： ii ’ ected surface in previously developed nvnverse w 出 下 er to ima 驴 s captured rom complex scene in chapter 4 ， an 《 dent ibr method is prese 庇 仙 y takn ull 讪 antage of 呷 bies c 咖 the method can simulate the 3d details on sllri ： ace of object successfully he 。 叩 proach ， called rered ature mopmp consists of two pans at fst ， an origi 。 ltexture with orthogonal displacements per pixel is deco 啊 osed into a series of new t6 刀 mfcs with each 他 lug a given displacement per pixel ， called ae ， ea atures ， or lt hen hese lt e used to render the novel view by conventional texture mapping d avoid gaps n the endered hlla 驴 ， some phels are to be interpolated nd extended in the 厂 kaccoding to the depth differe eee between two neighbor pixels in the original texture as these ltlt fc … e much storage nd therefore much time is equired to install ltlt into the text ’ ufc buffec an 旧 thod is pfoposed to co 呷 fcss the ltlt ， nd the cottcspondingfclldering method is given experimental esults show that the new method is efficient ， especially n rendering those objects with a smaller depth rnge compared withtheir size ， such as relief surfaces of building
与己有的三维变换方法相比较，该方法不但成功地填补了由于投影区域扩张而产生的第一类空洞，而且成功地填补了由于空间深度非连续物体相互遮挡而产生的第二类空洞，从而方便地实现了虚拟环境中的漫游；基于物体表面深度的连续性，本文提出了一个位移预测方法? ?此方法可以从单幅参考图象获得逆映射过程中所需要的目标图象的位移信息，从而大大提高了算法的效率：与通常的正向映射算法相比，此算法克服了多幅参考图象所带来的计算量成倍增长等问题，而且误差较小。 2 ）基于极线几何的快速逆映射算法。利用参考图象的边界信息与隐含的遮挡关系，以及极线几何的性质，本文第三章提出了一个基于极线几何的快速3 ‘一中国科学院软件研究所博士学位论文基于图象的快速绘制技术的研究逆映射算法，从多幅参考图象精确合成当前视点目标图象。 |
| 10. | About the applications of cognitive map : line feature detection and form based on fuzzy cognitive map , basic shapes recognition based on fuzzy cognitive map , object recognition based on probabilistic fuzzy cognitive map , a new method of image understanding based on cognitive map and robot high - level planning based on fuzzy cognitive map have been realized in this dissertation
对于认知图在图像分析及理解中的应用：本文在认知图理论研究的基础上，把认知图应用到图像分析与图像理解中。实现了一种基于模糊认知图的线特征检测与形成方法、基于模糊认知图的基本形状识别方法、基于概率模糊认知图的目标识别方法及基于认知图的图像理解方法，并将图像理解与认知图应用到机器人的高层规划系统中。 |
