The spatial pixel resolution of common multispectral and hyperspectral sensors is generally not sufficient to avoid that multiple elementary materials contribute to the observed spectrum of a single pixel. To alleviate this limitation, spectral unmixing is a by‐pass procedure which consists in decomposing the observed spectra associated with these mixed pixels into a set of component spectra, or endmembers, and a set of corresponding proportions, or abundances, that represent the proportion of each endmember in these pixels. In this study, a spectral unmixing technique is proposed to handle the challenging scenario of non‐linear mixtures. This algorithm relies on a dedicated implementation of multiple‐kernel learning using self‐organising map proposed as a solver for the non‐linear unmixing problem. Based on a priori knowledge of the endmember spectra, it aims at estimating their relative abundances without specifying the non‐linear model under consideration. It is compared to state‐of‐the‐art algorithms using synthetic yet realistic and real hyperspectral images. Results obtained from experiments conducted on synthetic and real hyperspectral images assess the potential and the effectiveness of this unmixing strategy. Finally, the relevance and potential parallel implementation of the proposed method is demonstrated.

Manual selection of features from massive unstructured point cloud data is a very time-consuming task that requires a considerable amount of human intervention. This work is motivated by the need of fast and simple algorithm to obtain robust, stable and well-localized interest points that are used for subsequent processing in computer vision real-time applications. This paper presents an algorithm for detection of interest points in three-dimensional (3D) point cloud data by using a combined 3D Sobel–Harris operator. The proposed algorithm is compared with six state-of-the-art approaches used to identify the true feature points. Extensive experiments were carried out using synthetic benchmark and real datasets. The datasets were selected with different sizes, features and scales. The results were evaluated against human generated ground truth and predefined feature points. Three measures were used to evaluate the algorithm accuracy, namely localization accuracy Le, False Positive Error (FPE) and False Negative Errors (FNE). Also, the complexity analysis of the proposed algorithm is presented. The results show that the proposed algorithm can identify the interest points with accepted accuracy. It works directly on point cloud datasets and shows superiority when compared with other methods work on 3D mesh data.

Watermarking is the art of hiding intellectual property information into digital media for the purpose of copyright protection. 3D laser scanned archeological models consume an essential effort to build up a final pruned 3D model. The model is then deployed online through insecure channel. Our objective is to watermark the 3D scanned archeological models by embedding the ownership information into it. One of the main challenges of mesh watermarking is to avoid the convergence problem. The convergence problem will be solved by our proposed Nearest Neighbor Move (NNM) technique. In NNM the vertex position will be changed slightly to meet a predefined relationship between the vertex position and its projection on the opposite edge in each triangle of the triangular mesh. Results showed that the perturbing of vertices is reduced compared to the previous methods, in addition the watermark is embedded only in the bumpy regions which are imperceptible by Human Visual System. NNM showed a comparatively enhancement in the mesh quality and the convergence problem is successfully solved.

3D objects are used in several applications such as animations, virtual reality and visualization. These applications often require a huge amount of data storage and transmission. Since most applications require compact storage, fast transmission, in this paper, we present a novel procedure for compression and coding of three-dimensional (3-D) semi-regular meshes using wavelet transform. This procedure is based on space frequency quantization (SFQ) which is used to minimize distortion error of the reconstructed mesh for a different bit-rate constraint. Our experiments show that the 3D SFQ code over performs progressive geometry coder (PGC) in terms of the quality of compressed meshes. Moreover, bit-stream can be truncated at any point and still decode reasonable visual quality meshes.

Navigation is playing an increasingly important role in virtual environments (VE). Today virtual worlds are very large and present challenging navigation tasks. Usually a VE contains few visual details and hence few landmarks to support navigation. In this paper, we propose a technique to automatic identify landmarks in large virtual environment using spatial data mining techniques. Identified landmarks are used to enrich landmark knowledge which is essential for navigation in VE. Initially three datasets with different number of objects have been generated based on a realistic distribution of buildings on one of newest and well designed Egyptian cities. A new model for extracting landmarks in VE was developed based on the models used in real environment. The proposed technique was applied on the generated cities and successfully identified the landmarks.