搜索资源列表
refpaper6_hcrnumkannada
- Abstract. This paper describes a system for isolated Kannada handwritten numerals recognition using image fusion method. Several digital images corresponding to each handwritten numeral are fused to generate patterns, which are stored in 8x8 ma
Learning-depth-information
- 本文提出一种基于高斯- 马尔科夫 随机场模型，首先通过图像采集及激光测距系统，采集大量图像及其相匹配的深度信息图，在 人类视觉系统基础上，提取图像特征，通过训练完善模型，并应用于新采集图像上-This paper presents a Gauss- Markov random field model, first by image acquisition and laser ranging system, collecting a large number of images to ma
flash-id-does-not-match
- 在使用FL2440时，编译bootloader会出现flash id does not match错误，这是解决方案。-When exploiting the FL2440, there will merge erro "flash id does not match" after compiled bootloader, this is the answer.
Match-Ontologies-on-the-Semantic-Web
- Learning to Match Ontologies on the Semantic Web
matched-impedance
- 怎样理解匹配阻抗，电子技术方面的，应该属于嵌入式的范围-Understand how to match the impedance of electronic technology, should belong to the scope of embedded
model-CANDIDE
- 在模型与人脸初步匹配后, 对模型局部进行网格优化,提高了模型的表征力同时并不影响匹配速度 -In the initial model and face after the match, the model for local mesh optimization to improve the characterization of the model at the same time does not affect the matching speed force
a
- 阵元失效会破坏拖曳线列阵的幅相分布，导致阵列的旁瓣级出现明显升高，严重影响了阵列的性能。阵元失效条件下 的波束形成是一个非线性的最小平方的优化问题，对其直接求解非常困难。针对这一情况，提出了一种基于遗传算法的阵元 失效校准方法。算法的基本思想是将失效阵元的权重强制为零，并使得阵列的实际响应与期望响应在主瓣区域相匹配，同时 对阵列响应的旁瓣级作出限制-Failed elements will des仃oy the amplitude and phase distribution and