In robotic applications and especially 3D map generation of indoor environments، analyzing RGB-D images have become a key problem. The mapping problem is one of the most important problems in creating autonomous mobile robots. Autonomous mobile robots are used in mine excavation، rescue missions in collapsed buildings and even planets’ exploration. Furthermore، indoor mapping is beneficial in finding and rescuing missions. With recent advances، mobile robots are used in hazardous missions such as radioactive areas or collapsing buildings. Having the environment’ s map beforehand can boost efficiency and effectiveness of the mission. In order to digitize the environment، several 3D scans are needed. However، these scans should be merged according to a global coordination system to create a correct، consistent model. This process is called image registration. If the robot with 3D scanner is able to accurately localize itself، the registration can be done directly by robots pose. However، due to imprecise robot sensors، self-localization is error prone. Therefore، the geometric structure of overlapping 3D scans is considered. In order to registering various points sets، Iterative Closest Point (ICP) algorithm is used. ICP is the most common approach to align point clouds in two consecutive image frames. This algorithm uses a point to point approach. RGB and depth images which are captured by Kinect are used in this study. In order to reducing data points and performing faster 3D map creation، depth images are converted to point clouds and then segmentation is done according to image planes. For this purpose RGB images are segmented by region growing segmentation algorithm. In this algorithm، the image was initially over segmented. This algorithm uses stack data structure and Euclidean distance in Lab color space to segment the image. Euclidean distance in Lab color space describes the resemblance of two colors to each other. In this algorithm، the aim is to label each pixel to a segment. To this end، each unlabeled pixels Euclidean distance to its neighboring mean color is checked to be within a threshold. For over-segmentation، if the distance satisfies the smaller threshold، the more pixels will be merged to the segment. Afterwards a plane was fit to each segment. After segmentation، each segment should be represented by a plane. Eventually، the segments were merged based on the product of normal vectors and plane fitting error criteria. After segmentation، planes were fit to the new segments again. A given number of points were generated on the plane. ICP algorithm was executed on these points and transfer and rotation matrices were obtained. Generating points on the plane results in fewer points. Therefore، the points were reduced and algorithms performance was increased. The results show that the proposed method increases the speed up to 55 and 91 percent in consecutive and non-consecutive frames on average، respectively.

Roland Barthes identifies two types of relationships between text and image: anchorage and relay. Anchorage helps identify one meaning in the image while it can be potentially the site of multi-farious meanings. The complementary relationship of text and image is called Relay. Furthermore, Kress and Leeuwen have suggested another relationship, called “one image-many words”, in which image is free to construct meaning. However, there are some other types of relations between image and words in a text, which have not been examined by Barthes or even Kress and Leeuwen or other individuals. That is, when an image leads to a metaphoric change in the interpretation of the text, simply the conceptual blending theory (Fauconnier and Turner) could be a proper method to explain how the human’s mind could understand and interpret a new meaning through the interaction of image and words (they said interaction of two different concepts).In the present paper, the way in which visual texts could be created and interpreted in the human’s mind as well as the different types of relations between text and image are studied based on four spaces, which are basic for meaning construction and four different types of blending pattern. Finally, a new relationship, which is not so common in any other texts other than literary one, will be introduced.This paper explores the tools, which are used in literary text to free the meaning from the control of words. It also studies the role of the reader in the interpretation of a text in visual poetry. Furthermore, it is going to argue how the interpretation of a text could be changed by the image, not the text.In conclusion, the image space (input space 1) could map its presupposed common features onto the feature of the words space (input space 2) and then project them into the blending space to create new meaning. So the given meaning could be developed from close meaning to far or metaphoric one through the active role of the reader in the interpretation of the floating signified.

Subsurface channels are stratigraphic features in seismic data that can act as reservoirs or conduits for hydrocarbons. However, detecting and characterizing these channels is challenging due to the limitations of seismic resolution and the complexity of the subsurface geology. Seismic inversion is a technique that can enhance the seismic data by transforming the seismic traces into quantitative estimates such as acoustic impedance (AI), which is a key reservoir rock property. AI inversion can help to identify and delineate the subsurface channels by providing more contrast and detail of the channel geometry, fill, and surrounding sediments. Seismic inversion is often challenged by the non-uniqueness, ambiguity and uncertainty of the inversion results due to noise and band-limited data. This paper uses a fuzzy model-based seismic inversion method that integrates prior information and fuzzy clustering constraints to produce more realistic and reliable AI models. This method assigns data points to multiple clusters with varying degrees of membership, which can capture the overlapping of AI values of different geological formations. The method is applied to the 3D Poseidon seismic data from the Browse Basin, offshore Western Australia, and the results are compared with those of conventional model-based inversion. Since there is no well-data in an interest channel zone, a qualitative evaluation with seismic attributes is performed. The subsurface structures are further interpreted by various seismic attributes. The comparison shows that the fuzzy model-based inversion method can improve the resolution, contrast and stability of the AI models and reveal more detail of the subsurface geology.

Spectral decomposition of time series has a significant role in seismic data processing and interpretations. Since the earth acts as a low-pass filter, it changes the frequency content of the passing seismic waves. Conventional methods of representing signals in a time domain and frequency domain cannot show the time information and the frequency information simultaneously. Time-frequency transforms an upgraded spectral decomposition to a new step and can show time and frequency information simultaneously.Time-frequency transforms generate a high volume of spectral components, which contain useful information about the reservoir and can be decomposed into single frequency volumes. These single frequency volumes can overload the limited space of a computer hard disk and are not easy for an interpreter to investigate them individually; therefore, it is important to use methods to decrease the volume without losing information. The frequency slices are thus separated from these volumes and used for an interpretation.In this study, three different methods were used to represent a buried channel. In the first method, the numbers of the single frequency slices were investigated, variations of the frequency amplitudes in the slices were observed, and an expert interpreter could obtain some information about the channel content and lateral variation. Since different frequencies contain different types of information (low frequencies are sensible to channel content and high frequencies are sensible to channel boundaries), none of the slices were able to show all information simultaneously. In the next two methods using a color stacking method, the RGB plots were constructed which, due to the different frequency content, resulted in more information than the frequency slice representation method.An RGB image, sometimes referred to as a true color image, is an image that defines red, green, and blue color components for each individual pixel and has an intensity between 0 and 1. In this study, RGB plots were constructed in two different manners, RGB plots based on conventional RGB plot methods and RGB plots using basis functions. In the conventional method, three different frequency slices were mapped against the red, green and blue components. Although this method obviates some drawbacks of the single frequency plots, it uses only three slices and practically ignores a big part of information. Using basis functions and defining windows, the interpreter was able to introduce some frequency intervals and plot them against the primary components and use the total bandwidth or its major part. Three simple raised cosine functions having different frequency centers and different periods were chosen. The image quality strongly depended on these two parameters. Longer window widths will introduce longer frequency widths into every primary component and resulted in smoother color combinations for images and very short periods had the same results as the conventional RGB plot method. Different centers showed different details. Low frequency centers showed channel content properties, and high frequency centers showed channel boundaries and fine branches.In this study, the spectral decomposition was first performed on land seismic data from an oil field in Iran using a short time Fourier (STFT) transform and an S transform. Then three demonstration methods were applied for channel detection. Finally it was shown that how RGB color stacking method represented buried channels in more precise images and how a basis function based RGB represents better results than the conventional RGB method.

Shortening of words in all languages of the world has had many reasons. The economic principal in language, quick transmission of the purpose, Poetic necessity and the use of artistic aspects of the language, can all be some of the different reasons for this purpose. There are different methods for shortening of words, among them abbreviation and blending are the most important ones. This article studies these two processes In shortening of words.

The purpose of this study is to scrutiny the political news headlines of Ebtekar newspaper based on the theory of conceptual blending. The Analysis of the theory is based on the emerging meaning and structure. This theory consists of four inputs, each of which plays a role in formation of the emergent meaning. Also in this theory, the types of blending networks (simple, mirror, single-scope, and double-scope) are introduced according to the format of inputs. This article examines the political headlines in Ebtekar newspaper to determine the adequacy of the theory in explaining and describing the headlines. For this purpose, 1000 news headlines have been analyzed. The results showed that about 20% of the headlines have used blending. In terms of the frequency of blending, the single-scope type is the most frequent one because this type of blending has four elements: simplicity, transparency, communication, and attractiveness. Therefore, by using this type of blending, in addition to conveying information, the author attempts to make the headlines more attractive, convince the audience and persuade them to follow the news. Also, mirror-type blending is not used in research data, because the structure of this type of blending is such that two events are formed in a common semantics frame but at two different times. Given this structure, it may not be easy for the reader to recall and match two events.