Recently, the use of micro-Doppler radar signatures for target classification has become an area of focus, in particular for the case of dynamic targets where many components are interacting over time. To fully exploit the signature information, individual scattering centers may be automatically extracted and associated over the full target observation. The availability of ultrafine radar range resolution, or micro-range resolution, aids this process immensely. This paper proposes one such algorithm. The proposed method uses the well-known nonlinear least squares (NLS) and expectation-maximization (EM) algorithms. As shown, leveraging fine range and Doppler resolution allows human signatures to be decomposed into the responses of constituent body parts. The algorithm is experimentally validated against a number of measured human-radar data sets.

Micro-Range/Micro-Doppler Decomposition of Human Radar Signatures

This paper proposes an automatic skin cancer (melanoma) classification system. The input for the prosed system is a collected data images, it followed by different image processing procedures to enhance the image properties. Two segmentation methods used to identify the normal skin cancer from malignant skin and to extract the useful information from these images that passed to the classifier for training and testing. The features used for classification is the coefficients created by Wavelet decompositions and simple wrapper curvelet. Curvelet is suitable for the image that contains oriented texture and cartoon edges. Recognition accuracy of the three layers back-propagation neural network classifier with wavelet is 51.1% and with curvelet is 75. 6% in digital images database.

The automatic identification of melanoma by wavelet and curvelet analysis: Study based on neural network classification

This paper proposes an automatic skin cancer (melanoma) classification system. The input for the proposed system is a set of images for benign and malignant skin lesions. Different image processing procedures such as smoothing and equalization are applied on these images to enhance their properties. Two segmentation methods are then used to identify the skin lesions before extracting the useful feature information from these images. This information is then passed to the classifier for training and testing. The features used for classification are coefficients created by Wavelet decompositions or simple wrapper Curvelets. Curvelets are known to be more suitable for the images that contain oriented textures and cartoon edges. The recognition accuracy obtained by the two layers back-propagation neural network classifier tested in this experiment is 58.44 % for the Wavelet based coefficients and 86.57 % for the Curvelet based ones.

/pdf/wavelet-and-curvelet-analysis-for-automatic-identification-3iwqpotep9.pdf

Wavelet and Curvelet Analysis for Automatic Identification of Melanoma Based on Neural Network Classification

This paper proposes an automated non-invasive system for skin cancer (melanoma) detection based on Support Vector Machine classification. The proposed system uses a number of features extracted from the Grey Level Co-occurrence Matrices (GLCM) of the gray-scale skin lesion images, and color features obtained from the original color images. The dataset used include digital images for skin lesions that are either benign or malignant. The purpose of this project is to provide a system that can classify digital images of skin lesions as benign or malignant (Melanoma). The testing accuracy obtained by the Support Vector Machine classifier used in this experiment is 82.7% for the GLCM features and 81.48% for both GLCM and Color features using ROI segmentation. The proposed system has resulted in a sensitivity of 83.6 % for the case of GLCM and 83.33% for the case of GLCM and color using ROI segmentation. It has also resulted in a specificity of 80% for the case of GLCM and 76.19% for the case of both GLCM and Color features using ROI segmentation. The obtained sensitivity and specificity results are comparable to those obtained by dermatologists. Consequently, this can increase the chance of the survival from Melanoma.

Automatic non-invasive recognition of melanoma using Support Vector Machines

This paper proposes an automated non-invasive system for skin cancer (melanoma) detection based on Support Vector Machine classification. The proposed system uses a number of features extracted from the Wavelet or the Curvelet decomposition of the grayscale skin lesion images and color features obtained from the original color images. The dataset used include both digital images and Dermoscopy images for skin lesions that are either benign or malignant. The recognition accuracy obtained by the Support Vector Machine classifier used in this experiment is 87.7.1% for the Wavelet based features and 83.6. 6% for the Curvelet based ones. The proposed system also resulted in a sensitivity of 86.4 % for the case of Wavelet and 76.9% for the case of Curvelet. It also resulted in a specificity of 88.1% for the case of Wavelet and 85.4% for the case of Curvelet. The obtained sensitivity and specificity results are comparable to those obtained by Dermatologists.

Automatic recognition of melanoma using Support Vector Machines: A study based on Wavelet, Curvelet and color features

Curvelet transform is one of the recently developed multiscale transform, which can well deal with the singularity of line and provides optimally sparse representation of images with edges. But now the image denoising based on curvelet transform is almost used the Monte Carlo threshold, it is not used the feature of images’ curvelet coefficients effectively, so the best result can not be reached. Meanwhile, the wavelet transform codes homogeneous areas better than the curvelet transform. In this paper a joint multiscale algorithm with auto-adapted Monte Carlo threshold is proposed. This algorithm is implemented by combining the wavelet transform and the fast discrete curvelet transform, in which the auto-adapted Monte Carlo threshold is used. Experimental results show that this method eliminate white Gaussian noise effectively, improves Peak Signal to Noise Ratio (PSNR) and realizes the balance between protecting image details and wiping off noise better.

A Joint Multiscale Algorithm with Auto-adapted Threshold for Image Denoising

Inverse Synthetic Imaging LIDAR:No Targets Can Hide

Micro-Doppler induced by mechanical vibrating or rotating of structures in a radar target is potentially useful for target detection, classification and recognition. While the Doppler frequencies induced by the target body scatterers are constants after motion compensation, the micro-Doppler due to vibrating or rotating structures of the target is a function of dwell time. Now the characteristic of micro-Doppler has been used in low speed moving targets' recognition widely, but high speed moving targets have different properties from the low speed moving targets. Due to the influence of high speed, the returned signal after stretch processing will appear range profile broadening and range migration. This paper analyzes the micro-Doppler signature of high speed moving targets, and computer simulations are given to prove the validity of the analysis.

Qun Zhang

Papers

A Joint Multiscale Algorithm with Auto-adapted Threshold for Image Denoising

Inverse Synthetic Imaging LIDAR:No Targets Can Hide

Micro-Doppler analysis of high speed moving targets

Inverse Synthetic Imaging LIDAR : No Targets Can Hide(WSANE 2009 (Workshop for Space, Aeronautical and Navigational Electronics))