Time filter

Source Type

Xiao B.,Chinese National Engineering Research Center for Information Technology in Agriculture | Guo X.,Chinese National Engineering Research Center for Information Technology in Agriculture | Lu S.,Open Key Laboratory of Information Technology in Agriculture | Wen W.,Open Key Laboratory of Information Technology in Agriculture | Qin Q.,Open Key Laboratory of Information Technology in Agriculture
ICIC Express Letters | Year: 2011

The example-based method is an effective approach for complex features modeling of plant. However, accurate and efficient example matching is a hard task for many applications. In this paper, an example matching and retrieval approach based on Biomimetic Pattern Recognition (BPR) is presented for virtual plant modeling. Firstly, an example database is founded by scanned plant models. Secondly, features of a modeling object are extracted and a relevant feature space in 3D space is constructed by hyper sausage neuron according to the features. And then, the similarity between object model and the example model is evaluated by Biomimetic Pattern Recognition and the most similar example is selected. By use of the example database, the proposed algorithm has been implemented and the experimental results show that the embedded watermark is effective and robust for example matching and retrieval. Finally, main contributions and limitations are discussed in conclusion. © ICIC International 2011.


Xiao B.,Chinese National Engineering Research Center for Information Technology in Agriculture | Wen W.,Chinese National Engineering Research Center for Information Technology in Agriculture | Guo X.,Open Key Laboratory of Information Technology in Agriculture | Lu S.,Chinese National Engineering Research Center for Information Technology in Agriculture
ICIC Express Letters, Part B: Applications | Year: 2011

Digital plant modeling is becoming a hot issue both in computer graphics and biologic research, and it is widely used in many fields of digital life such as virtual reality, film making, landscaping, amusement and games. A principal problem is high-precision modeling of various plants with many complex features. Development of 3D (three dimensional) digitization provides a novel way to acquire the true data of plants. In this paper, a digital plant colony modeling approach based on 3D digitization is proposed, and the method is implemented by a group of maize. Firstly, spatial data of maize colony are collected by a digitizer, and the data are organized by corresponding files. Furthermore, a cardinal based three-curve model is constructed based on original points; thus the maize colony model is expressed by several key parameters. Finally, an experimental system is implemented by C++ program language and rendered by OpenGL library. Experimental results show that the modeling method is effective for digital plant colony modeling. At the end, main advantages and limitations as well as the future works are discussed. © 2011 ICIC International.


Wang X.,Chinese National Engineering Research Center for Information Technology in Agriculture | Wang X.,Open Key Laboratory of Information Technology in Agriculture | Xiao B.,Chinese National Engineering Research Center for Information Technology in Agriculture | Xiao B.,Open Key Laboratory of Information Technology in Agriculture | And 6 more authors.
Journal of Information and Computational Science | Year: 2011

In this paper, a method based on motion retargeting for human-like character animation of maize plant driven by motion capture data is presented. Generally, plants have no articulated structures thus the character animation of plant object is a hard task. Firstly, the curve-skeleton of the 3D maize plant model is extracted according to its media-axis pattern. Secondly, the mapping relationship between the source human skeleton and the target character skeleton is constructed, and the virtual joints of the target character skeleton are located in terms of the human skeleton proportions. Provided with rich motion capture data, our method could convert motions between hierarchically and geometrically different characters and output the human-like animation of maize plant. The experimental results show that the proposed method is effective and robust. Finally, the main advantages and limitations are discussed. Copyright © 2011 Binary Information Press.


Xiao B.,Chinese National Engineering Research Center for Information Technology in Agriculture | Xiao B.,Open Key Laboratory of Information Technology in Agriculture | Guo X.,Chinese National Engineering Research Center for Information Technology in Agriculture | Guo X.,Open Key Laboratory of Information Technology in Agriculture | And 4 more authors.
Journal of Information and Computational Science | Year: 2010

Plant modeling is an attractive issue both in computer graphics and virtual reality, however, the complex features modeling is still a challenge for many researchers. In this paper an example-based modeling method for complex features of plant is presented including three steps. Firstly, the features are extracted by several parameters, and then the extracted features are used to find the similar examples in database by feature matching. The selected example is deformed according to the extracted features and a new model is generated. The experimental results show that the proposed approach is effective and robust for plant complex features modeling. Finally, the main contributions and limitations are discussed. Copyright © 2010 Binary Information Press.


Xiao B.,Chinese National Engineering Research Center for Information Technology in Agriculture | Xiao B.,Open Key Laboratory of Information Technology in Agriculture | Zhao C.,Chinese National Engineering Research Center for Information Technology in Agriculture | Zhao C.,Open Key Laboratory of Information Technology in Agriculture | And 6 more authors.
Journal of Computational Information Systems | Year: 2010

Template-based modeling is an effective approach for complex plants structures. A skeleton-based template retrieval method is presented for virtual maize modeling. The retrieval is performed based on a template database composed of many template models by three steps: affine transformation, curve discretization and curve matching. The main process of skeleton-based template retrieval method as well as assistant processing such as template generation is introduced in detail. To examine the algorithm, some experimental results are given and illustrated. The experiments show that the algorithm retrieves similar templates in database exactly and it is an effective and robust approach for template retrieval. Finally, main advantages and limitations are discussed. Copyright © 2010 Binary Information Press May, 2010.

Loading Open Key Laboratory of Information Technology in Agriculture collaborators
Loading Open Key Laboratory of Information Technology in Agriculture collaborators