- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
Investigation of Remote Sensing Image Fusion Strategy Applying PCA to Wavelet Packet Analysis Based on IHS Transform
摘要： Further exploration of wavelet packet analysis (WPA) in the area of image fusion has been a hot topic. It is a strategy to combine WPA with such other transforms as intensity–hue–saturation (IHS), principle component analysis (PCA) for image fusion between the panchromatic (PAN) and the multispectral (MS) image. The paper puts forward a distinct fusion method. Its main idea can be stated as three steps. Firstly, intensity component is derived from IHS model of the image after an MS image is transformed from RGB to IHS. Secondly, intensity component and a matched PAN image are decomposed by WPA at the second scale, respectively. The innovational concept with two aspects is applying PCA theory to merge wavelet packet coefficients. One is to detect edge and produce self-adaptive weighted ratios for low-frequency band. The other is to yield another weighted coefficients for high-frequency bands based on standard deviation. Lastly, the new intensity component created by implementing inverse WPA, matching with hue and saturation reserved, makes up a color composition. A fused image is produced when carrying out transformation from IHS to RGB for the composition. It turns out that the presented fusion strategy is effective with experiments.
关键词： Intensity–hue–saturation (IHS),Image fusion,PCA-based fusion rule,Principle component analysis (PCA),Wavelet packet analysis (WPA)
更新于2025-09-23 15:23:52
-
AIP Conference Proceedings [AIP Publishing PROCEEDINGS OF THE 2ND INTERNATIONAL CONFERENCE ON BIOSCIENCES AND MEDICAL ENGINEERING (ICBME2019): Towards innovative research and cross-disciplinary collaborations - Bali, Indonesia (11–12 April 2019)] PROCEEDINGS OF THE 2ND INTERNATIONAL CONFERENCE ON BIOSCIENCES AND MEDICAL ENGINEERING (ICBME2019): Towards innovative research and cross-disciplinary collaborations - Laser-induced breakdown spectroscopy (LIBS) for printing ink analysis coupled with principle component analysis (PCA)
摘要： Laser-induced breakdown spectroscopy (LIBS) has been applied to perform elemental analysis of printing ink samples. Samples of black printing inks from three types of printers viz. inkjet, laser-jet, and photocopier (three different brands for each type) and one control sample (blank white A4 paper) were analysed under optimised conditions. Results revealed that the LIBS method when coupled with PCA was able to provide discriminative evidence on elemental differences among all the different printing inks. Considering its time and cost effectiveness as well as requiring only minute amount of sample with no sample pre-treatment steps, the combination of LIBS and PCA may prove useful for forensic questioned document practical caseworks.
关键词： forensic questioned document,Laser-induced breakdown spectroscopy,principle component analysis,PCA,printing ink analysis,LIBS
更新于2025-09-16 10:30:52
-
[IEEE IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia, Spain (2018.7.22-2018.7.27)] IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Innovative Multi Pcnn Based Network for Green Area Monitoring - Identification and Description of Nearly Indistinguishable Areas - In Hyperspectral Satellite Images
摘要： The paper presents an original neural network approach for region of interest detection and classification in multi-spectral satellite images. The proposed method uses a sequence of Pulse Coupled Neural Networks that identifies plausible regions of interest. These regions are passed to a dimension reduction algorithm, Principle Component Analysis, in order to generate the input data for a Support Vector Machine classifier, that validates the data. The algorithm's parameters are optimized using a Genetic Algorithm. The algorithm is designed to distinguish regions that are extremely similar, such as parks in a city that has entire districts made up of houses with yards. The algorithm has been tested on images provided by the Sentinel-2 satellite, and it proved that it can recall 76.85% of the pixels marked as park in the ground truth data, which was obtained from OpenStreetMap.
关键词： Genetic Algorithm (GA),Pulse Coupled Neural Network (PCNN),Principle Component Analysis (PCA),Support Vector Machine (SVM)
更新于2025-09-10 09:29:36
-
[IEEE IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia, Spain (2018.7.22-2018.7.27)] IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Wind Field Retrieving for SCAT Onboard CFOSAT Based on PCA Method
摘要： The scatterometer (SCAT) onboard China France Oceanography Satellite (CFOSAT) will be the first scanning scatterometer with rotating fan-beams. Wind retrieving methods for existed scatterometers, which are with rotating pencil beams or fixed fan beams, are needed to be modified considering efficiency in information extraction from the numerous observations acquired in this working mode. This was achieved in the research of this paper by applying Principle Component Analysis (PCA) method. Firstly, vectors for applying PCA are composed in two different ways for analysis of the effectiveness of information extraction for the SCAT data and for wind retrieving respectively. Then the description of simulated SCAT data considering observing geometry and SCAT working mode was given. Then experiments of PCA based wind retrieving method was carried out with conclusion that it was effective in information extraction for preparing data sets for wind retrieving. Finally, further research has been discussed.
关键词： wind retrieval,principle component analysis (PCA),China-France Oceanography Satellite (CFOSAT),Scatterometer (SCAT),simulated data
更新于2025-09-09 09:28:46