CEDD:54字节色彩边缘导向描述符，高效图像检索

5星 · 超过95%的资源需积分: 9 119 浏览量更新于2024-09-09 收藏 382KB PDF 举报

CEDD (Color and Edge Directivity Descriptor) 是一种专为图像检索设计的高效、紧凑特征表示方法，它在图像处理领域具有重要意义。由 Savvas A. Chatzichristofis 和 Yiannis S. Boutalis 在2008年的 ICVS (International Conference on Visual Surveillance) 上提出，论文发表于 LNCS (Lecture Notes in Computer Science) 5008卷，页码312-322。该研究旨在解决大型图像数据库中的索引和检索问题，特别强调了在计算效率上的优势。 CEDD 的核心在于将颜色和纹理信息整合到一个直方图中，使其特征长度控制在每个图像仅为54字节，这使得它非常适合在大规模图像库中应用。相较于MPEG-7等其他标准的图像描述符，CEDD 的提取过程所需的计算资源显著减少，这在资源有限或对实时性能有高要求的场景下显得尤为关键。论文作者使用了一种名为ANMRR（Average Normalized Mutual Retrieval Rate）的客观评价指标来衡量CEDD 提议特征的性能。ANMRR 反映了特征在实际检索任务中的准确性和召回率，通过对比测试结果，能够直观地评估CEDD 在图像检索任务中的有效性。此外，论文还提供了一个在线演示，展示了如何将CEDD 特征集成到图像检索系统中，用户可以直观地体验到其在实际应用中的效果。这个演示无疑为研究人员和工程师提供了一个实践案例，以便他们更好地理解和利用这一高效的颜色和边缘方向描述符。 CEDD 是一个在图像检索领域具有竞争力的特征提取技术，它在压缩信息的同时保持了足够的区分度，适用于那些对计算效率和存储空间有严格要求的场景。对于图像检索系统的开发者和研究人员来说，理解和掌握CEDD 的原理与应用，将有助于优化图像搜索算法并提升系统的性能。

314 S.A. Chatzichristofis and Y.S. Boutalis

Fig. 1. Edges Extraction with CLF-AND Filter

membership values of Figure 2(a). The selected hue regions are stressed by dotted lines

in figure 1(d). The membership limit values of S and V are identified with the same

process.

The use of coordinate logic filters (CLF) [10] was found to be the most appropriate

among other edge detection techniques for determining the fine differences and fi-

nally extracting these vertical edges. In the procedure followed, each pixel is replaced

by the result of the coordinate logic filter “AND” operation on its 3x3 neighborhood.

The result of this action, stresses the edges of the image. The total edges are exported

by calculating the difference between the initial and the filtered image.

Based on these edges, the inputs of the system are analyzed as follows:

Hue is divided into 8 fuzzy areas. Their borders are shown in figure 2(a) and are

defined as: (0) Red to Orange, (1) Orange, (2) Yellow, (3) Green, (4) Cyan, (5) Blue,

(6) Magenta and (7) Blue to Red.

S is divided into 2 fuzzy regions as they appear in figure 2(b). This channel defines

the shade of a color based on white. The first area, in combination with the position of

the pixel in channel V, is used to define if the color is clear enough to be ranked in

one of the categories which are described in H histogram, or if it is a shade of white or

gray color.

The third input, channel V, is divided into 3 areas (Figure 2(c)). The first one is ac-

tually defining substantially when the pixel (block) will be black, independently from

the values that the other inputs have. The second fuzzy area, in combination with the

value of channel S gives the gray color.

A set of 20 TSK-like rules [11] with fuzzy antecedents and crisp consequents was

used. The consequent part contains the variables that count the number of the original

image pixels, which are mapped to each specific bin of the 10 bin histogram. Four of

the rules depend on two only inputs (S and V). For these rules the decision is inde-

pendent from the value of H. For the evaluation of the consequent variables two

methods have been used. Initially LOM (Largest of Maximum) algorithm was used.

This method assigns the input to the output bin which is defined from the rule that

gives the greater value of activation. Next, Multi Participate algorithm was tried. This

defuzzyfication method assigns the input to the output bins which are defined from all

the rules that are being activated. More details about this algorithm are available in

[6]. The experimental results show that the second algorithm performs better.

Next, a second system undertakes the task to separate each color in 3 hues. This

system forms a 24 bins histogram as an output. Each bin represents a preset color as

follows: (0) Black, (1) Grey, (2) White, (3) Dark Red, (4) Red, (5) Light Red, (6)

Dark Orange, (7) Orange, (8) Light Orange, (9) Dark Yellow, (10) Yellow, (11) Light

剩余10页未读，继续阅读

BecomeWhoWeAre

粉丝: 1
资源: 4

CEDD:54字节色彩边缘导向描述符，高效图像检索

基于Java的ES图片搜索插件开发

基于CEDD特征的图像搜索引擎设计与实现

使用LIRE实现图像相似度检测技术

CEDD: Color and Edge Directivity Descriptor.

星瞳视觉生成.pdf

一个面向大数据的智能图像搜索系统.pdf

55.香港启德机场Kai Tak地区规划设计_CEDD.zip

交通物流建筑与规划-55.香港启德机场KaiTak地区规划设计_CEDD

1b7fd969a199623d8d329ef0a34c2cedd5bc5558b741985b9109f752eb0f5421

up_510902_pcdtr_sixqb.rar

最新资源