Abstract
This paper describes a new type of global, gridded product being created by the Multi-angle Imaging SpectroRadiometer (MISR) team. The product is a compressed version, or summary, of MISR geophysical data products on a 1° monthly global grid. Data belonging to each grid cell are summarized by a multivariate histogram. The numbers, sizes, and shapes of the histogram bins vary among cells, and they adapt to the shape of the data in high-dimensional space. Also, bin representatives are means rather than midpoints. These modifications allow data to be summarized parsimoniously and with lower error than is possible using customary, simple, descriptive statistics. The method is demonstrated by compressing test MISR aerosol data, and performance is assessed by comparing computations using compressed data with those using the original.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 1626-1636 |
| Number of pages | 11 |
| Journal | IEEE Transactions on Geoscience and Remote Sensing |
| Volume | 40 |
| Issue number | 7 |
| DOIs | |
| State | Published - Jul 2002 |
Keywords
- Clustering algorithms
- Data compression
- Entropy-constrained vector quantization
- Level 3 products
- Massive datasets
ASJC Scopus subject areas
- Electrical and Electronic Engineering
- General Earth and Planetary Sciences