Work in the field of fire and smoke detection is becoming an increasingly covered subject. Conventional algorithms use exclusively models based on feature vectors. These vectors are difficult to define and depend largely on the type of fire being treated. These traditional methods give results with low detection rates and high false classification rates. The current trend is to take an innovative approach to solving this problem by using an algorithm to automatically determine useful features to classify fire and smoke. In this paper, we propose a convolutional neural network to identify fire and smoke from real-time images. Convolutional neural networks have shown their great performance in the field of object classification. Tested on real image sequences, the proposed approach achieves better classification performance than conventional methods. These results clearly indicate that the use of convolutional neural networks for fire detection is very encouraging.

In this paper, we propose an unsupervised classification scheme based on the Dempster-Shafer Theory (DST) and the Dezert-Smarandache Theory (DSmT) to characterize vegetated, aquatic and mineral surfaces. From pre-processed ASTER satellite images (georeferencing, geometric correction and 15 m re-sampling), neo-channels were produced by determining the spectral indices NDVI, MNDWI and NDBaI, considered as sources of information for classification of a given pixel. Then, we modeled respectively the formalisms of the DST and the DSmT and we realized the algorithms and related codes that we implemented in the MATLAB environment. Our contribution lies in taking into account the imperfections (inaccuracies and uncertainties) linked to source information through the use of mass functions based on a simple Gaussian distribution support model in order to model each focal element independently of the others and to evaluate the belonging of a pixel to a class with respect to the majority of elements representing said class. The resulting results show that the DST approach is relatively satisfactory for the unsupervised classification of mineral surfaces and aquatic surfaces while it is not satisfactory for vegetated surfaces according to all proposed models. As for the DSmT, it presents satisfactory results for all the models proposed. The model with the exclusion integrity constraint E (Intersection) V (Intersection) M= Phi was selected as the best model because having, in addition to an average rate of well-graded pixels of 93.34%, a compliance rate of 96, 37% with the terrain higher than those of the other models implemented.

In this paper, we propose a new representation of characteristics based on texture and color analysis for rock recognition. The proposed method combines the discriminating colour and texture characteristics of a rock image from a composite LBP descriptor to make automatic, fast and efficient rock identification. Indeed, the colorimetric texture descriptor ALBPCSF (Adjacent Local Binary Pattern based on Color Space Fusion) derives from the concatenation of the LBP texture characteristics and the color characteristics with the fusion of the two (02) colorimetric spaces RGB and HSV. In our methodology we first applied ALBPCSF on images of two (02) different families of rocks that are magmatic rocks and metamorphic rocks to produce colorimetric texture images then the K-SVD (K-Singular Value Decomposition) dictionary algorithm with a choice of suitable parameters is applied to said texture images produced to calculate a signature of the rocks from our image base. For dictionary learning the K-SVD method uses Orthogonal Matching Pursuit (OMP) as a sparse coefficient coding algorithm. The experimental results of the proposed approach on our image database show that the results of the proposed color LBP are relatively better than those with a grayscale or scalar LBP on the one hand and better than those of the direct K-SVD on the initial images on the other hand. The proposed strategy contributes significantly to improving the performance of automatic rock identification systems.