User Tools

Site Tools


point-in-polygon_20hit_20testing

This is an old revision of the document!


Point-in-Polygon hit testing

by Richard Russell, September 2010

A common requirement in graphics programming is to determine whether a point is inside or outside a polygon (which can be regular or irregular, concave or convex). Since any closed curve can be approximated by a polygon, given a sufficient number of vertices, this can be extended to testing whether a point is inside or outside an arbitrary boundary.

One approach is to use the Windows API. The polygon can be converted to a region using the CreatePolygonRgn API and the hit test then carried out using the PtInRegion API; however this is a relatively slow operation. This article presents an alternative method using an assembly language routine to perform the test directly.

To use this method the code must first be assembled, during the initialisation phase of your program, by calling PROCassemblepip as follows:

      PROCassemblepip

Once that has been done the test can be carried out simply by making the following SYS call:

      SYS `PtInPoly`, pt{(0)}, nvert%, x%, y% TO res%

This is directly equivalent to (but much faster than) this code using the Windows API:

      SYS "CreatePolygonRgn", pt{(0)}, nvert%, 1 TO hrgn%
      SYS "PtInRegion", hrgn%, x%, y% TO res%
      SYS "DeleteObject", hrgn%

In both cases pt{()} is an array of POINT structures containing the vertices of the polygon, nvert% is the total number of vertices in the polygon and x%,y% defines the point to be tested. The returned value res% is set non-zero if the point is inside the polygon and zero if it is outside.

The array of POINT structures can typically be created as follows (in this case for a hexagon):

      nvert% = 6
      DIM pt{(nvert%-1) x%,y%}

Once created the vertices should be initialised with the appropriate coordinates, for example:

      FOR V% = 0 TO nvert%-1
        READ pt{(V%)}.x%, pt{(V%)}.y%
      NEXT

Here the coordinates are assumed to be contained within DATA statements.

Finally, here is the code of PROCassemblepip:

      DEF PROCassemblepip
      LOCAL L%, P%, code%, pass%
      DIM code% 120, L% -1
      FOR pass% = 8 TO 10 STEP 2
        P% = code%
        [OPT pass%
        .`PtInPoly`
        mov edi,[esp+4]  ; lpPoints
        mov ecx,[esp+8]  ; nCount
        mov ebx,[esp+12] ; X
        mov ebp,[esp+16] ; Y
        mov esi,edi
        xor eax,eax
        dec ecx
        .piploop
        pushad
        mov ecx,[edi]
        mov edx,[edi+4]
        mov esi,[edi+8]
        mov edi,[edi+12]
        call intersect
        xor [esp+28],eax
        popad
        add edi,8
        loop piploop
        pushad
        mov ecx,[edi]
        mov edx,[edi+4]
        mov edi,[esi+4]
        mov esi,[esi]
        call intersect
        xor [esp+28],eax
        popad
        ret 16
        .intersect
        cmp edx,ebp  ; y1 > y0 ?
        setg al
        cmp edi,ebp  ; y2 > y0 ?
        setg ah
        xor al,ah    ; eor
        jz bailout
        cmp edx,edi  ; y1 > y2 ?
        setg al
        sub esi,ecx  ; dx0 = x2 - x1
        sub edi,edx  ; dy0 = y2 - y1
        sub ebx,ecx  ; dx2 = x0 - x1
        sub ebp,edx  ; dy2 = y0 - y1
        imul esi,ebp ; dx0 * dy2
        imul edi,ebx ; dy0 * dx2
        cmp esi,edi  ; > ?
        setg ah
        xor al,ah    ; eor
        .bailout
        movzx eax,al
        ret
        ]
      NEXT pass%
      ENDPROC
This website uses cookies for visitor traffic analysis. By using the website, you agree with storing the cookies on your computer.More information
point-in-polygon_20hit_20testing.1522502374.txt.gz · Last modified: 2018/03/31 13:19 by 127.0.0.1