《基于形状的叶片图像检索-外文翻译 (2篇)》由会员分享，可在线阅读，更多相关《基于形状的叶片图像检索-外文翻译 (2篇)（13页珍藏版）》请在金锄头文库上搜索。
1、外 文 翻 译毕业设计题目： 植物叶片的形状特征提取及分类研究 原文1： Shape based leaf image retrieval 译文1： 基于形状的叶片图像检索 原文2：Feature extraction and automatic recognition of plant leaf using artificial neural network 译文2： 利用人工神经网络实现植物叶片的特征提取与自动识别 Shape based leaf image retrievalZ. Wang, Z. Chi and D. FengAbstract: The authors prescnt
2、 an efficient two-stage approach for leaf image retrieval by using simple shape features including centroid-contour distance (CCD) curve, eccentricity and angle code histogram (ACH). In thc first stage, the images that are dissimilar with the query image will be first filtcred out by using ecccutric
3、icy to reduce the search space, and fine retrieval will follow by using all three sets of features in the reduced search space in the second stage. Different from eccentricity and ACH, the CCD curve is ncither scaling-invariant nor rotation-invariant.Therefore, normalisation is required for the CCD
4、curve to achieve scaling invariance, and starting point location is required to achieve rotation invariance with the similarity measure of CCD curves. A thinning-based method is proposed to locate starting points of leaf image contours, so that the approach used is more computationally cfficient. Ac
5、tually, the method can bencfit other shape representations that are sensitive to starting points by reducing the matching timc in image recognition and retrieval. Experimental results on 1400 leaf images from 140 plants show that the proposed approach can achieve a better retrieval performance than
6、both the curvature scale space (CSS) method and the modified Fourier descriptor (MFD) method. In addition, the twostage approach can achieve a performance comparable to an exhaustive search, but with a much reduced computational complexity.1 IntroductionPlant identification is a process resulting in
7、 the assignment of each individual plant to a descending series of related plant groups in ternis of their common characteristics.Such a task is very demanding in biology and agriculture, such as for the discovery of new species, plant resource surveys and plant species database management. So far,
8、this time-consuming process has mainly been carried out by botanists. A significant improvement can be expected if the plant identification can be carried out by a computer, automatically or semautomatically,assisted by image processing and computer vision techniques. By using a computer-aided plant
9、 identification system, non-professionals can also identify many plant species. This will promote an interest in studying plant taxonomy and ecology, and lift primary and secondary biology education standards at various levels, and promote the use of infonnation technology for modernising the managc
10、mcnt of botanical gardens,natural reserve parks and forest plantation.A computer plays three major roles in computer-aided plant identification: information storage and retrieval,automatic image and text information processing, and the use of machine learning techniques for deriving decision trees f
11、or plant identification. Fig. 1 shows the simplified block diagram that we proposed for a computer-aided living plant identification system. In identifying plant species, huinan beings will observe one or more of the following: the whole plant (form, size etc.), leaves (organisation,shape, margin, v
12、enation patterns etc.), flowers(inflorescence, growing position, colour, size, symmetry etc.), stem (shape, nodc, bark patterns, outer character etc.), fruits (size: colour, outer character, quality etc.).Some of these human observations can be carried out bya computer when the corresponding images
13、are provided.However, automatic (machine) plant recognition from colour images is still one of the most difficult tasks in computer vision, duc to: (i) lack of proper models or representations, (ii) a great number of biological variations that a species of plants can take, and (iii) imprecise image
14、prc-processing techniques such as edge detection andcontour extraction, thus resulting in possible missing features. Owing to many difficulties involved, research and development in computer-aided identification is still in its infancy.Plant leaves have an approximately two-dimensional nature an4 th
15、erefore, they are most suitable for machine processing. As the shape of plant leaves is one of the most important features for characterising various plants visually, the study of leaf image retrieval schemes will be an important stage for developing a plant identification system. By using such a le
16、af retrieval subsystem, a user can have a number of top-matched leaves together with their whole plant images and text descriptions displayed on the screen, which will help the user to further refine his/her identification by doing more observations.Im ef a/. I and Abbasi ef a/. 2 have done some preliminary work on plant recognition and classification with the s
