Informative accuracy investigation and updating map using remote sensing technique and GIS

In this work, using GPS which has best accuracy that can be established set of GCPs, also two satellite images can be used, first with high resolution QuickBird, and second has low resolution Landsat image and topographic maps with 1:100,000 and 1:250,000 scales. The implementing of these factors (GPS, two satellite images, different scales for topographic maps, and set of GCPs) can be applying. In this study, must be divided this work into two parts geometric accuracy and informative accuracy investigation. The first part is showing geometric correction for two satellite images and maps.The second part of the results is to demonstrate the features (how the features appearance) of topographic map or pictorial map (image map), Where it is highlight the different features with different scales to know the accuracy of information. Where can be noticed through graphics that features appear very close to each other at a certain scale and become detached at another scale and this problem can be solved by generalization method. Geodatabase in GIS program also have been used as a modern style store all data related to the project in one folder divisions with that data. In addition to storage in a shape file. Geographic Information Systems (GIS) uses Remote Sensing (RS) data for a lot of applications. One of the application areas is the updating of the GIS database using high resolution imagery. Finally, high resolution satellite imagery data is very important to obtain updating map for Kut city by implementation two methods geodatabase and shapefile.


Introduction
Maps are fast information sources when we want to know, to study and to develop projects in any region. Users, when consulting a map, expect information to be found. Sometimes, information can not appear due to scale or to the not updating of the map. Keates (1982) affirms "all maps are selective". In this way, it is necessary to adapt the scale and content to the desired objective of the map. In Brazil, most of the available maps to users are in an unsatisfactory scale and not updated.
Nowadays, we have developed different methodologies to updating maps, using especially orbital images from Landsat-TM and SPOTHRV. It is important to consider and evaluate questions such as area dimension, methodology quickness and easily, updating quality and change detection capacity through time. The detection capacity is associated to the informative content of the images used, whether they are aerial photographs or orbital images. This paper presents the continuation of results of the research done by Viadana (1995) where tests were performed with Landsat-TM and SPOT-P (panchromatic), showing the potential of orbital images to updating 1:50,000 topographic maps [1,2].
Being the blood of the geography information system (GIS), spatial data is the basement of the digital and information building. So the production of the spatial data never stops, and series database of different scales have been established. The currency of those databases would fade with time lapse if they cannot be updated timely. So the production and updating of the spatial data will be the central work all the time. With the help of GPS, GNSS and remote sensing (RS) image of high resolution, spatial data with good accuracy and currency can be acquired easily. Facing the increasing of big data, how to use them properly becomes a new problem. The technique of spatial data fusion is just the answer for the scientific use of existing data. Since different kinds of data are the abundant material of data fusion, then better data can be derived from the existing data through data fusion [3].

Study area and materials used 1. Description of study area
Al Kut is a city in eastern Iraq, on the left bank of the Tigris River, about 160 kilometers (99 miles) south east of Baghdad. It has the coordinates (32°30′20″N 45°49′29″E). It has an area of (17012km 2 ) which constitute 4% of total area of Iraq (441000km 2 ).   Table 2 location of study area. 5-ERDAS Imagine, ArcGIS and GPS.  Methodology of work 1. Steps of work 1-Scanning for the old topography maps with scales 1:250,000 and 1:100,000. 2-Geometric correction for these two maps.
3-Also applying geometric correction for Quick bird satellite image that has high resolution about 61cm by utilizing GPS to collect GCPs to study area, then analysing the errors showing by this technique. 4-Geometric correction is implemented for Landsat satellite image which has low resolution bout 30m using GPS to scurfy study area and analysing the errors producing by this process. 5-After that, ArcGIS are using to draw new digital map to get update map for study area. 6-Building geodatabase for ground features using ArcGIS.

Geometric correction
Digital images collected from airborne or space-born often contain systematic and non-systematic errors that arise from the earth curvature, platform motion, relief displacement, non-linarites in scanning motion, and the earth rotation. Some of these errors can be corrected by using ephemeris of the platform and precisely known internal sensor distortion characteristics. Other errors can only be corrected by matching image coordinates of physical features recorded on the image to the geographic coordinate of the same feature collected from a map [4].

Polynomials
A Polynomial is a mathematical expression consisting of variables and coefficients. A coefficient is a constant, which is multiplied by a variable in the expression, and the variables in polynomial expressions can be raised to exponents. The highest exponent in a polynomial. A polynomial with one variable, x, is given in the following form [5]: The following first-order polynomial transformation equation can be used to determine the required coefficients to transform pixel coordinate to the corresponding other coordinate value.
where (X,Y) are the input pixel coordinates and (X 0 , Y 0 ) are the output (geographic) coordinate. The order of the polynomial used in this process is the order of the transformation. Polynomial equations are used to convert the source coordinate to rectify the coordinate. The pixel coordinate system has an x coordinate (column) and y coordinate (row).The relationship between the pixel coordinate system and the geographic coordinate system can be defined by polynomial transformation. The best order of transformation can be obtained using a trial and error process.
Initially, a few (at least three for first-order polynomial) ground control point (GCPs) are required to determine six transform each set of row (X) and column (Y) pixel coordinate to output (Geographic) coordinates. In this work, second-order transformation equation (where at least 6 number of GCPs are required to determine 12 transformation coefficient) for X and Y are:-X 0 = a 1 + a 2 X + a 3 Y + a 4 X 2 + a 5 XY +a 6 Y 2 (3) Discusses the results 1. The first part of work (Geometric correction) Tables 1, 2 and Figs. 6, 7 represent geometrical correction for six GCPs for two different satellite images resolution, also this method applied for 1:100 000 map and 1:250 000 map to obtain corrected images and maps with coordinates UTM (WGS84) can be used for drawing features in ArcGIS software.

The second part of work (Drawing by ArcGIS using shape file and geodatabase)
The results in this work is to demonstrate the features (how the features appearance) of topographic map or pictorial map (image map), Where it is highlight the different features with different scales to know the accuracy of information.
Where we noticed through graphics that features very close to each other and become detached at a certain scale and this problem can be solved by generalization method and also have been used geodatabase in GIS program, as a modern style store all data related to the project in one folder divisions with those data. In addition to storage in a shape file. Finally been a process of modernization of the landmarks that have occurred on the ground and production updating topographical maps with the emergence of this form of analysis features. In order to draw land use map, the topography map and digital layers with the scale of 1.25000 created in ArcGIS 9.3 and also QuickBird and Landsat satellite images taken from Landsat Satellite are used. The resolution of a map is the accuracy with which the location and shape of map features can be depicted for a given map scale. In larger-scale map, the resolution of features more closely matches real-world features because the extent of reduction from ground to map is less. In smaller scale map, the map resolution diminishes because features must be smoothed and simplified, or not showed at all. And then comes the drawing with less obvious from topographical map [In the map with scale 1:100,000 most of these features would be omitted from these maps, these maps cover more area while retaining a reasonable level of detail]. While the road network drawn from low resolution satellite image, the roads merged and close with each. Showing Figs. 8-12.   Fig.8: The network of roads from high resolution image at scale 1:100,000.  Fig.9: Illustrated the network of roads from map at scale 1:100,000. Fig.10: Producing the network of roads from low resolution image at scale 1:100,000.  Fig.11: The network of roads map for high resolution image at 1:100,000 scale.   Fig.12: Illustrated the network of roads map from map at scale 1:100,000.

B-The results for 1:250,000 scale.
Small-scale maps (1:250,000) show large areas on single map sheets, but details are limited to major features, such as boundaries, parks, airports, major roads, railroads, and streams. When drawing the same road network scale of 1: 250,000, note that the features are converging to each other more in the high-resolution image and that is almost be appearing as a single block in the topographic map (1:100,000), but in the low-resolution image be reduced to half, see Figs. 13-18. Fig.13: Illustrated the network for high resolution image with 1:250,000 scale.   Fig.14: Illustrated the network for map with 1:250,000 scale.   Fig.15: Presented the network for low resolution image with 1:250,000 scale.  Fig.16: The network road map for high resolution image with 1:250,000 scale.   Fig.17: The network of roads map from (map at 1:100,000) producing with 1:250,000 scale.

D-Digital Update map for Kut city from high resolution satellite image which has 1:24,000 scales drawing as layers and using geodatabase.
In the geometric accuracy and feature compilation assessment results show that the high resolution image can be used in scale (1:24,000) mapping processes and can identifying of small features. Fig. 21, showing the update map of Kut city from high resolution image, this map producing many layers which represented smaller areas. The resolution of a map is the accuracy with the location and shape of map features can be depicted for a given map scale. These Figs. 21, 22, showing the update map for kut from high resolution satellite image by using shapfile and geodatabase respectively. E-Digital update map for Kut city from topographic map at scale 1:100,000 and low resolution satellite image which has 1:24,000 scales drawing as layers.
In the Fig. 24 satellite image with low resolution been invoked, which previously corrected and the GeoDatabase drawn from corrected map with scale (1:100000) invoked as well. And as the low resolution image and the map corrected previously by using geometric correction in (Erdas) program, the image and map are of the same coordinates system UTM(wgs84) which is the system used in the study area (Kut city), that mean the drawn layers from corrected map coincide the updated low resolution satellite image. After that features are drown, which was not clear in the map of sale (1:100000) depending on updated satellite image which covered study area.
The Fig. 23 is obtained after invoking layers of the GeoDatabase drawn from map of scale (1:100000), and the later projected on updated low resolution image, and after drawning features which was not found in old map, the satellite image been hidden and only layers drawn from old map with scale (1:100000) remained. Also new layers obtained from satellite image, where these layers formed an updated map to the study area, it contains enough number of features and layers to the study area, with coordinates system UTM (wgs84). After that I symbolized the map layers such as streets, rivers, and residential area. And put legend represent (GeoDatabase), scale bar, north arrow, and title on the map. These map production have been compared within each other by using ArcGIS software and compiled features have been controlled in field.
In the comparison studies, firstly, the numbers of features in text, line, polygon and point layers have been detected, controlled and compared. In this stage, high resolution satellite image can be selected as the best because the feature numbers of the compiled data are more than the others. And the feature layers have been evaluated in detail.

Conclusions
1-In general, the best location of GCPs led to reduce the errors and increasing the accuracy. 2-For high resolution satellite image, when using the set of GCPs which is producing a closed errors in both x and y axes, while the root mean square error (RMSE) is decreased (0.5) and showing high accuracy. 3-The low resolution Landsat satellite image shows high RMSE (0.77) and low accuracy. 4-Database (maps) should be updated every time to give the real changes in study area. 5-A new updated digital topographic maps of the study area (Alkut city) have been obtained by using satellite images of different resolutions and maps of different scales (old maps) as a reference maps can be applied.
6-By using ArcGIS can be replace shape files database with feature class database (geodatabase) to ease invoking and transfer data from hard disc to hard disc and from computer to computer.