The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B7. Beijing 2008 
Figure 4.The final detection results 
4. CONCLUTION AND FUTURE WORK 
This paper proposes a fast method to detect object on large 
remote sensing image based on geostatical and local cluster 
analysis, and mainly focuses on ship detecting task on aerial 
image, which can be applied in many fast object detection fields 
such as harbor runtime management. The original data is 
sampled to a relatively smaller image and a LISA static is 
calculated for each kernel which centers on each pixel of each 
band of the sampled image. A novel static S is conducted to 
evaluate the extent how a kernel differs from the image back 
ground by combining LISA value of the kernel on three image 
bands and a threshold of S is set to extract the most different 
kernels we call anomalies which suggest the location of ships. 
A simple recognition process proceeds based on the location 
results. At last an image shot by airborne digital frame camera 
on East-See is experimented and all the 14 ships are properly 
detected. This algorithm is proved effective and timesaving. 
The future work will focus on how to extract objects on 
complex background images. Besides, the recognition method 
need researched more. 
Chong Jinsong Zhu Minhui,” Comparison on Ship Target 
Detection Algorithms of SAR Imagery”,Singnal Processing, 
6(2006). 
Wang Min and Luo Jaincheng and Ming Dongping,’’Extract 
Ship Targets from High Spatial Resolution Remote Sensed 
Imagery with Shap Featrue”,Geomatics and Information 
Science of Wuhan University, Aug. 2005,Vol. 30 No.8. 
Chang,C.I. and Chang, S.S. Anomaly detection and 
classification for hyperspectral imagery. IEEE Transactions on 
Geoscience and Remote Sensing, 40((2002)), 1314- 1325. 
REFERENCES 
Pierre Goovaertsa and Geoffrey M. Jacqueza and Andrew 
Marcus : “Geostatistical and local cluster analysis of high 
resolution hyperspectral imagery for detection of anomalies”, 
Remote Sensing of Environment, 95 (2005) ,351-367. 
Anselin, L., “Local indicators of spatial association-LISA”. 
Geographical Analysis, 27(1995), 93- 115. 
Goovaerts,P, “Geostatistical incorporation of spatial coordinates 
intosupervised classification of hyperspectral data”. Journal of 
Geographical Systems, 4 (2002), 99-111. 
McGrath,D. and Zhang,C.”Spatial distribution of soil 
organiccarbon concentrations in grassland of Ireland”.Applied 
Geochemistry, 18(2003), 1629- 1639. 
Xu Da-qi and Ni Guo-qiang and Xu Ting-fa, “Study on the 
algorithm for automatic plane classification from remote 
sensing images with mid- high resolution”, Optical Technique, 
Nov. 2006 ,Vol. 32 No. 6.