60 likes | 178 Views
Homework I. 2014-3-3. Experiment Setting. Image database UKBench dataset: http://pan.baidu.com/s/1jGgFY6q 10200 images from 2550 categories Feature extraction HSV color feature Quantization step: 2×2×9. Conversion from RGB to HSV. Tasks. Retrieval performance evaluation
E N D
Homework I 2014-3-3
Experiment Setting • Image database • UKBenchdataset: http://pan.baidu.com/s/1jGgFY6q • 10200 images from 2550 categories • Feature extraction • HSV color feature • Quantization step: 2×2×9
Tasks • Retrieval performance evaluation • Take each image as a query • Linearly search all database images • Performance measure by N-S score • Check the precision at top-4 returned results • Count how many of the 4 images which are top-4 when using a query image from that set of four images. • Compute the average performance for all queries • Performance measured by mAP • Compute the average time cost per image
Implementation • Program with C++ or Matlab • You may need Open CV when programing with C++ • OpenCV 210 library files are provided • You do not need to install the OpenCV source file • Refer to OpenCV China for instructions to set programming environment • http://wiki.opencv.org.cn/index.php/Template:Install • Refer to OpenCV China for instructions to process images : • http://wiki.opencv.org.cn/index.php/%E9%A6%96%E9%A1%B5