Application of texture features and machine learning methods to grains segmentation in rock material images

The segmentation of rock grains on images depicting bulk rock materials is considered. The rocks material images are transformed by selected texture operators, to obtain a set of features describing them. The first order features, second-order features, run-length matrix, grey tone difference matrix, and Laws' energies are used for that purpose. The features are classified using k-nearest neighbours, support vector machines, and artificial neural networks classifiers. The results show that the border of rocks grains can be determined with above 70% accuracy. The multi-texture approach was also investigated, leading to an increase in accuracy to over 77% for early-fusion of features. Attempts were made to reduce feature space dimensionality by manually picking features as well as by use of the principal component analysis. The outcomes showed a significant decrease in accuracy. The obtained results have been visually compared with the ground truth. The observed compliance can be considered satisfactory.

Identifying Medicinal Plant Leaves using Textures and Optimal Colour Spaces Channel

This paper presents an automated medicinal plant leaf identification system. The Colour Texture analysis of the leaves is done using the statistical, the Grey Tone Spatial Dependency Matrix(GTSDM) and the Local Binary Pattern(LBP) based features with 20 different colour spaces(RGB, XYZ, CMY, YIQ, YUV, $YC_{b}C_{r}$, YES, $U^{*}V^{*}W^{*}$, $L^{*}a^{*}b^{*}$, $L^{*}u^{*}v^{*}$, lms, $l\alpha\beta$, $I_{1} I_{2} I_{3}$, HSV, HSI, IHLS, IHS, TSL, LSLM and KLT). Classification of the medicinal plant is carried out with 70\% of the dataset in training set and 30\% in the test set. The classification performance is analysed with Stochastic Gradient Descent(SGD), kNearest Neighbour(kNN), Support Vector Machines based on Radial basis function kernel(SVM-RBF), Linear Discriminant Analysis(LDA) and Quadratic Discriminant Analysis(QDA) classifiers. Results of classification on a dataset of 250 leaf images belonging to five different species of plants show the identification rate of 98.7 \%. The results certainly show better identification due to the use of YUV, $L^{*}a^{*}b^{*}$ and HSV colour spaces.

False Color and Infrared Imaging for the Identification of Pigments in Paintings

ABSTRACTIn the study of cultural heritage, most of the analytical techniques are point-specific or give information about small areas of the object. Therefore it is essential to obtain an overview of which points are suitable for these further investigations. To fulfill this, a first imaging study is the best way to proceed. Hereby, we can record the entire piece at once and observe the behavior and relation between different materials of the object. Various types of light can be used to obtain a selection of images and consequently also different information about the artifacts. Among them, infrared (IR) photography can be used as a first analysis, for instance, to reveal the pigments’ response upon interaction with IR radiation.In following we will present results obtained via IR video-photography on a selection of painted objects from the Mexican cultural heritage. These items are analyzed by False Color procedure, where colors are assigned to every grey tone of the pure IR photo. Hereby it is possible to distinguish between certain pigments on the painted surface.

The Extraction of Song Brocade's Characteristic Elements and the Application in Fashion Design

In order to inherit and develop the valuable heritage of Chinese traditional clothing cultural-Song Jin, the characteristic elements, patterns and colors of Song Jin, were extracted. Summarized that the common themes of Song Jin were geometry, flowers, animals, utensils and meteorological theme and the Four Dayun, Six Dayun and Eight Dayun of geometry themes were the most classical patterns among all these themes; the common color combinations were divided into three categories. The first was the combination of harmonic colors which based on the yellow tone. The second was the combination of harmonic colors which based on the grey or blue-grey tone and used the reddish violet and blue as ornaments. The third often used the combination of contrasting colors which usually chose bright red, green and orange to highlight the main prints. On this basis, explore how to use the two characteristic elements in fashion design.

Effect of CeO2 and Sb2O3 on the phase transformation and optical properties of photostable titanium dioxide

In the present work, the effect of individual additives calculated as molar fractions of Sb2O3 and CeO2 (x Sb 2O3 range: 0.03–0.08 %, x CeO 2 range: 0.05–0.14 %), on the phase composition, phase transformation, and optical properties of photostable rutile titanium dioxide was studied using selective leaching method, ICP-AES technique, XRD method, spectrophotometric analysis and S BET measurements. The starting material was hydrated titanium dioxide. It was observed that the addition of Sb2O3 to TiO2 did not influence the anatase-rutile phase transformation, but increasing the CeO2 addition caused a decrease in the rutilization degree. Thus, CeO2 acted as an inhibitor of the TiO2 phase transformation. Sb2O3 addition to TiO2 presumably caused the formation of a co-phase of Sb with Ti. Cerium formed a separate phase, CeO2, and reacted partly with titanium, probably creating co-phase, Ce0.8Ti0.2O2. Comparing the colour of modified rutile titanium dioxide according to the type of the additive introduced, it was found that TiO2 with CeO2 had higher brightness but lower white tone values when compared with TiO2 modified with Sb2O3. The relative lightening power and grey tone of the modified TiO2 were higher in TiO2 modified with Sb2O3. The values of the photocatalytic activity measured in all TiO2 samples modified either with Sb2O3 or CeO2 were very similar and varied around the value of 21.