View source for AIS visualization TASK

== Introduction ==

<gallery>
Map1.png|The map and some coordinates
Map2.png|The map shall be transformed to photo
</gallery>

The position of vessels on a photo.

== Data acquisition  ==

USE RTL-AIS, ship162 (https://github.com/xoolive/ship162) project and Python, with tape measure antenna.

Or GNU AIS saves the data to local MySQL server.

== Transform ==

Some ideas to generate the transformation. Full camera projection is not needed here. 


The latitude/longitude scale is not a linear measure. Thus, to a conversion to metric scale is better. UTM projection (pyproj) or using a local tangent plane. The UTM projection works better for this 10-20 km distances.

=== Plan ===

The coordinate system of the map is WGS84 EPS:4326 to EPSG:32635.

The mapping and ideas
# Get the mapping ship's GNSS coordinates to image coordinates
## Convert GNSS coordinates to UTM and then to metric points. Change the origin such that the metric numbers are reasonable.
## Find the corresponding pixels at the photo image.

=== UTM projection ===

The latitude/longitude scale is not a linear measure. Thus, to a conversion to metric scale is better. UTM projection (pyproj) or using a local tangent plane. The UTM projection works better for this 10-20 km distances.

<syntaxhighlight lang="python">
# WGS84 GPS -> UTM zone 35N
transformer = Transformer.from_crs(
    "EPSG:4326",
    "EPSG:32635",
    always_xy=True
)


gps_points = np.array([
[59.4480442,24.7727679],
[59.4720746,24.7267166],
[59.5183884,24.7962952],
[59.5226077,24.5841614],
[59.5770763,24.7294084]
])

utm_points = []
for lon, lat in gps_points:
    x, y = transformer.transform(lon, lat)
    utm_points.append([x, y])

x0, y0 = utm_points[2]
local_points = []
for x, y in utm_points:
    local_points.append([x - x0, y - y0])

print(local_points)
</syntaxhighlight>

===  Homology ===

https://docs.opencv.org/4.x/d9/dab/tutorial_homography.html

The simplest;
<math>
\begin{bmatrix}
u\\ v \\ 1
\end{bmatrix}
\tilde{=} H
\begin{bmatrix}
x\\ y\\ 1 
\end{bmatrix}
</math>

Should have some 10-50 corresponding pairs to get the mapping correct. Use least squares or SVD, but OpenCV works with RANSAC.

=== Non rigid spatial warp ===

Thin plate spline, 
<math>
(u,v) = f(x,y)
</math>
where 
<math>
f(x,y) = a_1 + a_2 x + a_3 z + \sum_i w_i U(|p-p_i|)
</math>
and 
<math>
U(r) = r^2 \log r 
</math>

===  ===