I have to make a heatmap for Android Google Maps. I have geographical locations and numbers that are negative and positive weights which I want to visually represent. Unlike the majority of Himachal, I want to interfere with these positive and negative weight destructive; That is, when the two points are closer to one another and one is positive and the other is negative, the overlap of them interferes in a destructive way, effectively does not present those areas which are completely canceled We do. I am planning to use Android Google Map's Tile Overlay / Tile Provider category, on which to build / render the tile based on a given location and zoom. (I have no problem with this part.)
How should I go about presenting these tiles? I am planning to use the graphics class of Java, but the best I can think of is running through every pixel, it calculates what color it will be based on the surrounding data points Should and should project that pixel though it is very inefficient, though, and I was looking for suggestions on a better approach.
EDIT: I have considered everything to use a tile overlay to use a non-Android Google map inside the web-view. What a ground overlay I'm considering now I have a large 2-dimensional array of "squares". Each class has a long, latitude, and total +/- load, when a new data point is added, instead of providing it in the same way, it will be added to the "class" which is that which is. The weight of this data point will be added to the class and then I will use GoogleMap to present the square on the map polygon object The ratio is clear from 1 point, 1: blue (cold point), and & lt; With the ratio of having 1 red (hot point), points are numbered for the properties, which is determined by the color. Edit: Clustering data in small regional groups
I'm trying < Blockquote>
is going through every pixel, it calculates what color it should be based on the surrounding data points, and that pixel should be interpreted.
Even if it is slow, it will work, there are not too many All this is still done by translating it into Edit (moved from comment): In the edit you tell that clustering on LatLng I can not say that this is a better or a bad idea, but try something Is worthy. tiles on the screen, each
does not have too many pixels in the tile And all this is done on the background thread.
bitmap in
byte [] .
Byte [] is a representation of PNG or JPG file, so it is not a simple pixel mapping from
bitmap . The last operation takes some time and possibly more processing power than your entire algorithm may be required.
No comments:
Post a Comment