An Important Landmarks Construction for a GIS-Map based on Indexing of Dolly Images

Abstract

<p><em> In this paper, we describe the construction of important landmarks of roads in the GIS environment. The system uses the corners between more than two roads as an important landmarks. In this corner points will be saving a number of images, each one represents the movement direction between two segment roads. The objective of our work is to build the geo-database repository depend on the GIS (vector data) and multimedia (raster data) information. This paper considered as a preprocessing step for a roadmap discovery of video film when the input to the system is a dolly video film of roads with absence the GPS locations of frames video, and the output is the pathway (route) on the map inside the GIS environment. This work considered as a starting point for multimedia query inside the GIS environment. That’s mean when query the GIS system about a particular image or video, the GIS system must be able to know and determine the location of this image or video file on the map. Thus, the first step for doing this process needs to building and constructing an appropriate database for matching process later.</em></p>

Categorizing and measurement satellite image processing of fire in the forest greece using remote sensing

“…Now are a portion of the the whole thing that have been researched andconsidered in this scope. In [14][15][16], describe the scope harms and dangers in turkey especially through Aegean and Mediterranean areas caused by the fires of Forest. Numerous digital image processing methods utlized to study the wildfire categorizing and compute the wildfire influenced on the burned areas.…”

Detecting abnormal movement of driver's head based on spatial-temporal features of video using deep neural network DNN

“…DDDS involves extracting salient and important features and then tracking these features to make an accurate classification about drowsiness status. Videos consist of spatial-temporal information which is presented as temporal features over the consecutive frames as well as the spatial features represented by each frame (still image) [1,2] and the obtained trajectory from head tracking in a video is formed from a sequence of pixels in consecutive frames [3,4] 345 methods can be categorized into two types: spatial stamp methods; where the decision makes based on spatial feature of one frame at a time; and time stamp methods based on a sequence of frames at a time to represent the dynamic changes in the features over the time called spatiotemporal feature [5,6]. The robust and efficient classification methods depend on extracting salient and efficient spatiotemporal features that can be used in decision making as well as the possibility to deal with changing in luminance and resolution [7,8].…”