Image Processing Reference
In-Depth Information
The Cell at Subnanometer Resolution—The cell is the most basic building block
of life. Cell function is a direct reflection of cell structure, and we understand the
function of the cell only to the degree that we understand its molecular architecture.
The molecular defects that underlie human disease are rooted in changes in the
molecular components in the cell that alter cell structure and function. Molecular
medicine would be greatly advanced by a molecule level view of the cell since it
would guide the engineering of molecular therapies to fix what is broken in cells
in the disease state.
28.1.3 Temporal Scalability
Sensemaking often involves temporal reasoning, and may require handling data at
different time scales. For example, it may be necessary to understand long-term
patterns by looking at data over a period of years or even decades, and simultane-
ously understand near-term effects by looking at data over a period of hours or less.
Moreover, it may be necessary to integrate and perform correlative analysis on data
collected at different temporal scales based on acquisition technology. For instance,
in understanding fundamental principals of rain formation in clouds, it can be nec-
essary to integrate data collected 1000 times per second with data collected every
several minutes (radar data) and this information may then be fed into climate models
that work on the scale of years and decades.
28.2 Variety of Data
Our ability to collect data is increasing at a faster rate than our ability to analyze it [
3
].
Scientists, engineers, and analysts are often overwhelmed with massive amounts of
data from multiple sources and where the important information content exists in a
fewpieces. Therefore, we need to create newmethods to allow them to visually exam-
ine this massive, multi-dimensional, multi-source, time varying information stream
to make decisions more efficiently. The various data types include the following:
Textual data—Massive textual data from documents, speeches, e-mails, or web
pages now influence the problem domain. This data can be truly massive, contain
billions of items per day, and much of it must be analyzed in a time-critical matter.
Databases—Many corporate and government entities have constructed huge
databases containing a wealth of information. We require new algorithms for the
efficient discovery of previously unknown patterns in these large databases.
Geospatial data—Consider the data collected by satellites that image the earth.
We now have satellites that can create images at less than 1 m resolution and that
can collectively image the land surface of the planet in a very short time. These
images must be examined in a time-critical matter.
Search WWH ::
Custom Search