Tools
Sorted by:
Try your query at:
 |
 |
 |
 |
 |
 |
Results 21 - 30
of
5,499
Table III. Eigenvalues of covariance of high dimensional remotely sensed data. Eigenvalue Proportion Accumulation
1993
Table 1. Remotely sensed estimatesof wheat area andyield in the Yaqui Valley for survey years compared with official statistics. Estimated Official Estimated Official
"... In PAGE 2: ..., 1992). ( Table1 ). In addition, yields for individual fields provided by Here we used Landsat Enhanced Thematic Mapper local farmers were compared with the average of remote- Plus (ETMH11001) data, with 30-m spatial resolution, to esti- sensing estimates for pixels completely contained within their fields.... In PAGE 5: ...35 (not shown). Nonethe- yields achieved in 2001 ( Table1 ). Higher yield potential, less, of interest here is the explanatory power of the in turn, increases demand for fertilizer N because the model, which equaled 51% with nine variables.... ..."
Table 1: Subsidence velocities for the selected sites. In all cases SAR data of the European Remote Sensing Satellites ERS-1 and ERS-2 were used.
"... In PAGE 4: ...EXAMPLES Four sites characterized by different displacement velocities were selected to investigate the performance of differential SAR interferometry for land subsidence monitoring: the Ruhrgebiet (Germany), Mexico City (Mexico), Bologna (Italy), and the Euganean Geothermal Basin (Italy). The approximate subsidence velocities, the monitoring interval selected, the number of interferograms used and the expected estimation error are summarized in Table1 . In the following the results achieved for the four sites will be summarized.... ..."
Table 1: Comparison of PSNR, obtained MAXAD, guaranteed MAXAD and bit rates obtained on eight-bit grey value images (Lena, ultrasound image, remote sensing image)
in I
2002
"... In PAGE 2: ... Food.0008 Table1 : Texture classes froin VisTex database [6] used in classification expcriments Gr;rss.... ..."
Table 1. Remote sensing systems relevant to fire detection and monitoring VIS-MIR, visible, mid-infrared; TIR, thermal infrared Sensor and additional web resources Temporal Spatial resolution VIS-MIR bands TIR bands
"... In PAGE 2: ... 1999). Space and airborne sensors have been used to assess environmental conditions before and during fires and to detect changes in post-fire spectral response ( Table1 ). Remotely sensed data have been used to detect active fires (Roy et al.... In PAGE 6: ... Following a brief description of the available satellite sensor systems, this paper will provide a review of how remotely sensed imagery has been used to monitor and evaluate these fire descriptors. Remote sensing instruments and platforms Many different sensor platforms and instruments have been used to remotely map and monitor active fire characteris- tics and post-fire effects ( Table1 ). In terms of the remote sensing of active fire characteristics and post-fire effects, we can divide the available sensor systems into passive or active and then further into aerial or satellite sensors.... ..."
Table 4.2: USGS Landuse/Land-cover Classification System for Use with Remote Sensing Data Modified for the National land Cover Data (NLCD) and NOAA Coastal Change Analysis Program (NOAA, 2004; Jensen, 2004)
2006
Table 3: Lot area estimates by data-driven decomposition and eld-based classi cation for the Zeewolde study site. The sub-column \main class quot; shows the dominant cover type according to a remote sensing expert. The super-column su xes \incl. quot; and \excl. quot; indicate whether soil was included in or excluded from the set of possible components of boundary structures during classi cation. The unit of all values is the area of a single pixel.
"... In PAGE 14: ... The narrow boundary structures present were assumed to be composed of water, forest, grass, soil, and asphalt; therefore, each of these classes was characterised by upto three di erent distributions extracted from the image by a remote sensing expert in order to serve as endmembers for data-driven decomposition and eld-based classi cation. Application of these methods gave the results shown in the second and third super-column of Table3 . The threshold setting with which these results were obtained was chosen such that eA was minimal.... In PAGE 16: ...lassi cation. The unit of all values is the area of a single pixel. structures composed of soil, which probably did not even exist in reality. The results of eld- based classi cation after excluding soil as a possible boundary component are presented in the last columns of Table3 ; the results of data-driven decomposition are not shown as they deteriorated over the entire line. As expected, eld-based classi cation underestimated the area of most lots covered by soil less severely or even overestimated them; unfortunately, the area of two lots (numbers 15 and 17) whose main ground cover types were not soil was (further) overestimated as well.... ..."
Table 4. Transition Matrices for the protected and non-protected portions of the study area. Matrix row and column labels are the land cover types determined from remote sensing data. Each element in the matrix is the calculated state transition frequency for the Landsat scene from 1972 to 1988. Diagonal elements are retention frequencies. A. For an area of 695 Km2, outside the Reservation area of the Pinelands New Jersey. B. For an area of 1524 Km2, inside the Reservation area of the Pinelands New Jersey.
"... In PAGE 10: ... Comparison of transition frequencies inside and outside the protected area provided a quantitative measure of the effects of recent human activities in the Pinelands. Disturbances, indicated by transi- tions from mixed deciduous forest and Pine forest categories to non-forested were 15 to 26% higher outside the protected area ( Table4 ). Pine forest land cover presented a 72% retention frequency within the reserve versus only 44% outside the Reserve.... ..."
Table 1 shows which classifiers are currently parallelized on theHIVE and which are not. We are investigating the applicability (speed, classification accuracy, etc.) of those algorithms that have not been applied for classification of remotely sensed data. After this process, we will choose to parallelize some or all of the set below. All of the algorithms have been chosen because they show promise as unsupervised/supervised image classifiers.
"... In PAGE 4: ... Table1 : Status (S = Serial, P = Parallel code available) and applicability (M = applicable to Multispectral, H = applicable to Hyperspectral) of classification algorithms both supervised and unsupervised. Of the supervised classification algorithms that have been parallelized, common features of their performance on theHIVE stand out.... ..."
Table 3: Offset times for ight segments for RF01-RF03. RL denotes remote sensing leg; CB, cloud base leg; SC, subcloud leg; CT, cloud top leg; SP, special pattern (which varied from ight to ight); SF, surface ux leg. For the pro les we label full pro les (FP), cloud pro les (CP) and inversion pro les (IP).
"... In PAGE 36: ... The radar was operated during all ights. The radar legs (RL in Table3 ) own not far above cloud top yielded full coverage from the ocean surface to the top of the cloud. Flight segments in and below the cloud layer provided partial views.... ..."
Results 21 - 30
of
5,499
