In traditional IHS transformation, the panchromatic image directly replaces the intensity component, spatial information
of the panchromatic image is reserved. But it may cause severe spectral distortion at the same time. Enlightened by
correlation coefficient of two images and its physical meaning, a novel IHS transformation image fusion algorithm is
proposed. It's called local correlation coefficient weighted IHS transformation image fusion algorithm (LCCW-IHS).
The weighted parameter is determined by the local correlation coefficient between the high-resolution panchromatic
image and multi-spectral image's intensity component. Then the two images are fused and the new intensity component
is generated. Finally the fusion image is obtained by inverse IHS transformation. This method furthest synthesized the
region characteristics in the original images to be fused. Both the spectral characteristics of multi-spectral image and the
high- resolution features of the panchromatic image are maintained. And the texture details are also enhanced. The
experimental results of multi-spectral image fusion, analyzed by both subjective and objective evaluations, show the
proposed algorithm is effective for image fusion.
Most of the feature extraction methods is complex and is easily influenced by some uncertain factors. Based on the idea of multi-resolution decomposing theory, a new statistical feature extraction algorithm based on gray multi-channel decomposition is proposed in the paper. The image's gray is decomposed into some channels. The gray of an image is partitioned into some gray regions, which is also called gray channels. The gray features on each channel are analyzed and extracted. Then all the gray features form feature vector. The multi-channel feature extraction method based on multi-resolution decomposition using wavelet theory is introduced at first. Each channel's gray feature is extracted and formed the feature vector. Targets can be well distinguished. But the method not has invariant property when image rotates. Referenced the thought of multi-resolution decomposition, a new multi-channel feature extraction based on gray is proposed in the paper. The four features, which is average of gray, variance, number of pixel and peakedness, are extracted and form the feature vector. The vector is used to identify targets. With the new method, the shortage of the former method is overcome. And the method not only has invariant property of rotation, but also has invariant properties of proportion and translation. The robustness is perfect. And the calculation is simple. The features of some infrared images of tank and helicopter are extracted with the new method. The results show its effectivity. It's helpful for image feature extraction and target recognition.
The lateral inhibition mechanism of biologic vision is introduced and applied on edge extraction of images. With the
method, the image's original characters are unchanged. Furthermore, the edge extraction and image enhancement can be
done effectively when the gray changes caused by varying illumination. For increasing the matching ability of resistance
to geometric distortion, a new matching algorithm is established. The edges of the real-time and reference maps are
extracted with the above-mentioned method based on lateral inhibition. Then the concentric circle characteristic vector of
image is defined and the method of vector extraction is proposed. The concentric circle characteristic vectors of the real-time
and reference maps are extracted. Finally they are matched according to the vectors. The resistance of geometric
distortion is improved. A map of roadway in a city obtained by satellite is simulated. The results show that the influence
of gray and geometric distortion on scene matching is effectively overcome. The algorithm is easily implemented with
hardware. The operation speed of the algorithm is also fast. It's worth for the design of real-time scene matching system.
Support vector machine (SVM) can effectively improve the algorithm's generalization capability. This paper proposes a support vector machine target recognition method based on the target infrared feature and edge invariant moments. The simulation result shows it is faster than others. The discrimination is higher than the K-nearby method's one.
Based on the classical log-polar transformation (LPT), the variable parameter log-polar transformation (VPLPT) is proposed. In the new algorithm, the size of uniform sampling central area is freely adjustable according to different vision task and its high resolution is reserved. The peripheral area is processed with log-polar transformation, which base can be adjusted. The calculation quantity of the two algorithms is analyzed respectively and compared with each other. The result shows the new algorithm's rapidity.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.