Measuring distance between points

Hey,
I am trying to do 2-D particle tracking. What I am trying to do is for each point in the image, I want to find the shortest distance to anouther point. Repeat for each point. Then move on to the next image and then compare those distances, since the particles are moving slowly, to help keep tabs on the particle. So for example if it was at least 15 pixels away from the nearest particle in picture one, that should stay about the same in picture two. However I am having trouble trying figuring out how to find the closest particle to every particle. I am open to other suggestions as well for 2-D particle tracking.\

hi there,
i wrote a little vi (see attachment), but ist not tested.
best regards
chris
Best regards
chris
CL(A)Dly bending G-Force with LabVIEW
famous last words: "oh my god, it is full of stars!"
Attachments:
Points_LV71.vi ‏34 KB

Similar Messages

How to measure distance between two points uisng uiaccelerometer

Hello all,
I am trying to measure distance between two points.So for that i am used uiaccelerometer but its give only rotation changes. I am moving my whole device from one point to another point so for that all x,y & z changes remain same. So how can get the device movement for that?
Thank you..

UIAccelerometer does not give rotation changes, it senses acceleration in each of the 3 axis in g-force units. Moving in a plane from one point to another and stopping will result in a net g-force in that axis of zero. To get distance one has to measure the initial acceleration and then the time before a deceleration is detected. It gets really complicated in real life since the start and stop are not instantaneous.

Accurate distance between points, lat/long to miles?

Hi,
I have a bunch of points as lat/long data in SRID 8307 format. From reading this forum, I understand than in Oracle 8.1.7 to get accurate distances I need to transform these points into a cartesian coordinate system.
My data is US-based, so I am using SRID 32775 in a command like the following:
EXECUTE SDO_CS.TRANSFORM_LAYER('restaurant_locations', 'location', 'restaurant_locations_32775', 32775);
This creates a new table with new point geometries and a rowid that I assume points back to the original 8307 table.
I've tried creating an index on the new table with cartesian coordinates, but I get this error:
CREATE INDEX restaurant_csp_idx
ON restaurant_locations_32775(geometry)
INDEXTYPE IS MDSYS.SPATIAL_INDEX
PARAMETERS('SDO_LEVEL=9 sdo_commit_interval=1000 layer_gtype=POINT' );
2 3 4 CREATE INDEX restaurant_csp_idx
ERROR at line 1:
ORA-29855: error occurred in the execution of ODCIINDEXCREATE routine
ORA-13200: internal error [POINT] in spatial indexing.
ORA-29400: data cartridge error
ORA-13003: the specified range for a dimension is invalid
ORA-06512: at "MDSYS.MD", line 1673
ORA-06512: at line 1
ORA-13003: the specified range for a dimension is invalid
ORA-06512: at "MDSYS.MD", line 1673
ORA-06512: at line 1
ORA-06512: at "MDSYS.SDO_INDEX_METHOD", line 8
ORA-06512: at line 1
Here's the entry for the 32775-transformed points in the metadata table:
INSERT INTO USER_SDO_GEOM_METADATA
VALUES (
'restaurant_locations_32775',
'geometry',
MDSYS.SDO_DIM_ARRAY( -- 20X20 grid, virtually zero tolerance
MDSYS.SDO_DIM_ELEMENT('X', 1951908.05, 16230214.8, 0.005),
MDSYS.SDO_DIM_ELEMENT('Y', -6858801, 13168375.5, 0.005)
32775 -- SRID (reserved for future Spatial releases)
My questions are;
Do I need to build a new spatial index? It seems like once I transform the lat/long data to cartesian I need to build a new index as well ( on the 32775-transformed table ).
Is this the best way to approach distance queries with lat/long data? It seems like a lot of work, plus the second index and table really add to the overhead if a location changes.
Any ideas on why I can't build an index on the output table from my SDO_CS.TRANSFORM_LAYER() call? I used SDO_TUNE.ESTIMATE_TILING_LEVEL() and SDO_GEOM.VALIDATE_GEOMETRY() and got no complaints. I'm at a loss.
I also can't seem to get set autotrace to work. It works fine for any non-spatial query, but if I try to trace a spatial query, I get this error:
SQL> SELECT /*+ INDEX(restaurant_locations restaurant_sp_idx) */ r_a.restaurant_id
FROM restaurant_locations r_a, restaurant_locations r_b, user_sdo_geom_metadata m
WHERE r_b.restaurant_id != r_a.restaurant_id
AND SDO_GEOM.WITHIN_DISTANCE(r_a.location, m.diminfo, 1, r_b.location, m.diminfo) = 'TRUE'
AND r_b.restaurant_id = '5999';
2 3 4 5
RESTAURANT_ID
456999
456999
Execution Plan
ERROR:
ORA-01031: insufficient privileges
SP2-0612: Error generating AUTOTRACE EXPLAIN report
Statistics
49 recursive calls
28 db block gets
83 consistent gets
0 physical reads
0 redo size
415 bytes sent via SQL*Net to client
425 bytes received via SQL*Net from client
2 SQL*Net roundtrips to/from client
11 sorts (memory)
0 sorts (disk)
2 rows processed
I've looked at the arraysize, and I've made sure to run the trace-enabling sql and granted plustrace to my DB user.
Thanks for any help,
_jason

Hi Jason,
The error on the index create is likely due to data being outside the bounds of the coordinate system as specified in user_sdo_geom_metadata.
If the data is stored in the point field then you can check the bounds by doing queries like the following, altering them for your table/column name (my table is cities_test, the geometry column name is location):
SQL> select min(a.location.sdo_point.x) from cities_test a;
MIN(A.LOCATION.SDO_POINT.X)
-157.80423
SQL> select max(a.location.sdo_point.x) from cities_test a;
MAX(A.LOCATION.SDO_POINT.X)
-71.017892
SQL> select max(a.location.sdo_point.y) from cities_test a;
MAX(A.LOCATION.SDO_POINT.Y)
61.178368
SQL> select min(a.location.sdo_point.y) from cities_test a;
MIN(A.LOCATION.SDO_POINT.Y)
21.31725
Do you need to build a spatial index?
Only if you are going to use spatial operators such as sdo_filter, sdo_relate, sdo_within_distance, and sdo_nn. If you have no requirements for these operators, then there is no reason to build a spatial index. From the trace query at the end of the posting, I suspect that you will need to have a spatial index.
Is this the best approach? Maybe, it depends on what your requirements are. If the data is static and performance is your highest priority, then maybe it is. If you have a requirement for a spatial index, then certainly it is. If you are only getting the distance between a few few known geometries, and you don't care about the time it takes to convert data on the fly, then you can use the sdo_cs.transform function within the sdo_geom.sdo_distance function to convert both geometries to the equal area projection.
The validation routines should have caught this - I checked and they do not for 8.1.7, and they do for 9i.
Regarding the set autotrace command, I'm not sure why it isn't working for you. It works for my generic scott/tiger account from a typical install.
hope some of this is useful.
dan
null

MInimun Distance between points

Hi all, can you help me understand this program so that I can started:
public static int[] minDistance(int[] x,
int[] y)
Given N points in the plane with integer coordinates, determine the minimum distance between two distinct points in the input. The points will be given by two arrays of the same length, one containing the x-coordinates and the other containing the y-coordinates. Return the indices in the array of a pair of points at minimum (positive) distance from each other, least index first. If more than one pair of points have the minimum distance from each other, return the lexicographically first pair (i,j) of indices of closest points.
Constraints The two arrays will have the same length N, which will be at least two. The numbers in the input arrays will be between -10000 and 10000. Note that you are not guaranteed that there will be no duplicates in the list, but the requirements state that only positive distances are compared.
Examples
(1) { 0,1}, {0,1}. This input represents the two points (0,0) and (1,1). Return {0,1}.
(2) { 0,1,0}, {0,1,0}. Return {0,1} as before, not {0,2}. The point (0,0) occurs twice in the input list, but this does not cause the minimum distance to be zero.
(3) {0,1,2,3,4,5,6,7,8,9,10}, {0,12,34,-48,15,347,8,5,3,2,1}. Return {8,9}; the points with those indices are (8,3) and (9,2).
(3a) {0,1,2,3,4,5,6,9,8,7,10}, {0,12,34,-48,15,347,8,2,3,5,1}. Return {7,8}
Thanks..........

BTW, what's Sedgewick?He's an author.
http://www.cs.princeton.edu/~rs/
A little googling on my part reminds me that it was "divide and conquer" that he was talking about (too many) years ago.
http://www.facweb.iitkgp.ernet.in/~arijit/courses/autumn2006/cs60001/lec-compugeo-2.pdf

Distance between Points?

Am I missing the built-in method to determine the distance (float) between two CGPoints?
I know the formula
xd = x2-x1
yd = y2-y1
Distance = SquareRoot(xd*xd + yd*yd)
but I could not find a method in the docs.

I think you want the standard old C function sqrt()

Customizing distance between data Points on x-Axis

Hi,
I want to draw a LineChart.
I have these Timestamps [84, 1000, 34000, 34699, 439999] who
are represanting the x-Value of DataPoints along the X-Axis.
Unfortunately the distance between 2 datapoints along the
x-Axis is always the same, that means that between the points with
x-values 84 and 1000 is the same distance along the axis as between
the points with x-values 34699 and 439999.
But the distance between points with x-values of 34699 and
439999 should be much greater than between 84 and 1000.
How can I customize the distance between data Points on a
LineChart to solve my Problem?
I really dont know right now!
Greeting,
Z.

"zidaninho" <[email protected]> wrote in
message
news:gls479$nkt$[email protected]..
> Hi,
>
> I want to draw a LineChart.
> I have these Timestamps [84, 1000, 34000, 34699, 439999]
who are
> represanting
> the x-Value of DataPoints along the X-Axis.
>
> Unfortunately the distance between 2 datapoints along
the x-Axis is always
> the
> same, that means that between the points with x-values
84 and 1000 is the
> same
> distance along the axis as between the points with
x-values 34699 and
> 439999.
> But the distance between points with x-values of 34699
and 439999 should
> be
> much greater than between 84 and 1000.
>
> How can I customize the distance between data Points on
a LineChart to
> solve
> my Problem?
>
> I really dont know right now!
What happens if you convert the time stamps to Dates and use
a DateAxis?

Set reference point in objects and measure distance

1) How can I set the reference point in objects?
for example:
        a) exactly in the center of a pad or a via
        b) exactly in any corner of the board outline
2) How can I measure the distance of two points with the maximum accuracy?
  for example:
        a) measure the distance between two pads or via
        b) measure the lenght of a line

dear Semaphore, thanx for your response.
I already use the ways that you refer me, but like this I can't measure with accuracy, or setting the zero reference is not guarantee that I set it exactly in the center of via or pad.
With grid seted in smallest option is difficult to target the center of a pad, and also is difficult to target the point in the edje of a line with accuracy. With ruler I also have the same problem, I cant be sure that the ruler is pointing exactly to the center of a via or pad.
That I need is a kind of a snap option to attach the mouse-cursor exactly in the center of an object, that can help me to target exactly in the center and have an accurate measurement or accurate set the zero reference without mistake.
Any other suggestion for accurate measurements?

Calculating distance between Lat Long points

I'm wondering if anyone out there has done this before in LabVIEW. Does anyone have a model of the "Great Circle" calculation? I have a file of lat/long points that I need to calculate the distance between. Any help??
V/r,
Chris

I haven't done it in LV, but if you go to Wikipedia and search for Great Circle Distance you'll find the formual and an example.

Find driving distance between two points without using API by use of Lat & Long?

Using Google geocode API : http://maps.googleapis.com/maps/api/geocode/xml?address=thane&sensor=true
We performed get distance between search criteria entered by user and all related clubs by lat & long stored at db.
2. Two different points such as
(origin: Lat1 & Long1) and (destination: Lat2 & Long2)
We tried for to get distance between these two points,
(Lat2 & Long2) to (Lat1 & Long1)
But distance which we get by calculation is simple straight line distance
Origin Destination
(Lat1 & Long1) (Lat2 & Long2)
3. This is not driving distance as google shows in exact Km
4. For that Google provide another API (distancematrix API)
http://maps.googleapis.com/maps/api/distancematrix/xml?origins=Thane&sensor=true&destinations=khopat&mode=driving&language=en%20-%20EN
5. But there is limit for DistanceMatrix-Service without ClientID and client key
100 elements per query.
100 elements per 10 seconds.
2 500 elements per 24 hour period.
But as element request exceeds it shows : OVER_QUERY_LIMIT error
6. In case of Client ID and Client key
In Distance Matrix 100 000 elements per 24 hour period,a maximum of 625 elements per query and a maximum of 1 000 elements per 10 seconds.
As per this one there is option to get purchase these API but basic question is remain same for us if we are requesting single origin and multiple destination then how element calculation done by google?
But in document google says :
Elements
The information about each origin-destination pairing is returned in an element entry. An element contains the following fields:
Status: See Status Codes for a list of possible status codes.
Duration: The duration of this route, expressed in seconds (the value field) and as text. The textual representation is localized according to
the query's language parameter.
Distance: The total distance of this route, expressed in meters (value) and as text. The textual value uses the unit system specified with the
unit parameter of the original request, or the origin's region.

Any information that you see in a google map webpage can be retrieved using the API. The best way of finding the tags on the webpage is to manually perform the query using an IE webpage. Then capture the source and save to a file so you
can use a text editor to look at results. I often on a webpage use the menu : View - Source and then copy the source to a text file.
jdweng

Distance between all points

Hi all, I am trying to find the distance between all points in an array. I was able to get the all possible combinations but i am not getting the expected ouput.
For eg, if i have four points, total number combinations are going to be 6 and i want to find distance in all six combinations.
Please help me where i did mistake. Please find the attached vi.
Thanks
uday,
Please Mark the solution as accepted if your problem is solved and help author by clicking on kudoes
Certified LabVIEW Associate Developer (CLAD) Using LV13
Solved!
Go to Solution.
Attachments:
Find distance between all points.vi ‏17 KB

Try something like this.
Since I don't have the IMAQ function, I am using complex data for each point, but it would be trivial to adapt it to your data structures instead.
LabVIEW Champion . Do more with less code and in less time .
Attachments:
Find distance between all points.png ‏33 KB
Find distance between all pointsMODCA.vi ‏10 KB
Find distance between all pointsFP.png ‏21 KB

Distance between two GPS points

Is there an inbuilt function in PL/SQL for calulating the distance between two GPS (lat/long) points?
I'm using Oracle 9i. There's a thing called SDO_GEOM available, but I'm not sure if this is what its used for or if it's the best option.

If the earth is a complete globe and its circumference is 40000km,
I make the function as follows.
create ore replace
function distance
(a_lat number,a_lon number,b_lat number, b_lon number)
return number is
circum number := 40000; -- kilometers
pai number := acos(-1);
a_nx number;
a_ny number;
a_nz number;
b_nx number;
b_ny number;
b_nz number;
inner_product number;
begin
if (a_lat=b_lat) and (a_lon=b_lon) then
return 0;
else
a_nx := cos(a_lat*pai/180) * cos(a_lon*pai/180);
a_ny := cos(a_lat*pai/180) * sin(a_lon*pai/180);
a_nz := sin(a_lat*pai/180);
b_nx := cos(b_lat*pai/180) * co s(b_lon*pai/180);
b_ny := cos(b_lat*pai/180) * sin(b_lon*pai/180);
b_nz := sin(b_lat*pai/180);
inner_product := a_nx*b_nx + a_ny*b_ny + a_nz*b_nz;
if inner_product > 1 then
return 0;
else
return (circum*acos(inner_product))/(2*pai);
end if;
end if;
end;
I rewrite it by using the factorization and the triangle function's sum and difference formulas:
cos(x-y) = cos(x)cos(y)+sin(x)sin(y)
As result, this function is same to the 1st method of Billy Verreynne.
create or replace
function distance
(a_lat number,a_lon number,b_lat number, b_lon number)
return number is
circum number := 40000; -- kilometers
pai number := acos(-1);
dz number;
dx2y2 number;
inner_product number;
begin
if (a_lat=b_lat) and (a_lon=b_lon) then
return 0;
else
dz := sin(a_lat*pai/180)*sin(b_lat*pai/180);
dx2y2 := cos(a_lat*pai/180)*cos(b_lat*pai/180)*cos((a_lon-b_lon)*pai/180);
inner_product := dz*dz + dx2y2;
if inner_product > 1 then
return 0;
else
return (circum*acos(inner_product))/(2*pai);
end if;
end if;
end;
Message was edited by:
ushitaki

Excel formula to calculate the distance between multiple points using lat/lon coordinates

I'm currently drawing up a mock database schema with two tables: Booking and Waypoint.
Booking stores the taxi booking information.
Waypoint stores the pickup and drop off points during the journey, along with the lat lon position. Each sequence is a stop in the journey.
How would I calculate the distance between the different stops in each journey (using the lat/lon data) in Excel?
Is there a way to programmatically define this in Excel, i.e. so that a formula can be placed into the mileagecolumn
(Booking table),
lookup the matching sequence (via bookingId)
for that journey in the Waypointtable
and return a result?
Example 1:
A journey with 2 stops:
1 1 1 MK4 4FL, 2, Levens Hall Drive, Westcroft, Milton Keynes 52.002529 -0.797623
2 1 2 MK2 2RD, 55, Westfield Road, Bletchley, Milton Keynes 51.992571 -0.72753
4.1 miles according to Google, entry made in mileage column
in Booking table
where id
= 1
Example 2:
A journey with 3 stops:
6 3 1 MK7 7DT, 2, Spearmint Close, Walnut Tree, Milton Keynes 52.017486 -0.690113
7 3 2 MK18 1JL, H S B C, Market Hill, Buckingham 52.000674 -0.987062
8 3 1 MK17 0FE, 1, Maids Close, Mursley, Milton Keynes 52.040622 -0.759417
27.7 miles according to Google, entry made in mileage column
in Booking table
where id
= 3
I understand that 100% accuracy is not possible, so it will not be an issue.

http://www.cpearson.com/excel/LatLong.aspx
and
http://www.contextures.com/excellatitudelongitude.html

Distance between two points with degrees and minutes

I would like to store several points in the database given degrees and minutes as position. In this example I have point 1 that is E 150, 0/S 30, 0 and points 2 that is E 150, 0/S 30.1. For example if I enter 2 km as distance from position of point 1 I would like the search to return all points witin 2 km.(should return points 2 that is very near)
If I run the query
SELECT c.name
FROM cola_markets_cs c
WHERE
SDO_WITHIN_DISTANCE(c.shape,
SDO_GEOMETRY(2001, 8307, SDO_POINT_TYPE(150.0, 30.1, NULL), NULL, NULL),
'distance=10000 unit=KM')
= 'TRUE';
I get result both position 1 and 2, but if I decrease to distance=1000 no rows is returned.
Can you see what I am doing wrong here?
Can I also combine SDO_WITHIN_DISTANCE with SDO_NN_DISTANCE so I can ask for all points 10 km in distance from the reference point and I can also see the actual distance for each point?
Thank you
I have tested with the following code. Do I save the position wrong..?
CREATE TABLE cola_markets_cs (
mkt_id NUMBER PRIMARY KEY,
name VARCHAR2(32),
shape MDSYS.SDO_GEOMETRY);
INSERT INTO cola_markets_cs VALUES (
1,
'Point 1',
MDSYS.SDO_GEOMETRY(
2001,
8307,
MDSYS.SDO_POINT_TYPE(150.0, 30.0, NULL),
NULL,
NULL
INSERT INTO cola_markets_cs VALUES (
2,
'Point 2',
MDSYS.SDO_GEOMETRY(
2001,
8307,
MDSYS.SDO_POINT_TYPE(150.0, 30.1, NULL),
NULL,
NULL
-- UPDATE METADATA VIEW --
-- Update the USER_SDO_GEOM_METADATA view. This is required
-- before the Spatial index can be created. Do this only once for each
-- layer (i.e., table-column combination; here: cola_markets_cs and shape).
INSERT INTO USER_SDO_GEOM_METADATA
VALUES (
'cola_markets_cs',
'shape',
MDSYS.SDO_DIM_ARRAY(
MDSYS.SDO_DIM_ELEMENT('Longitude', -180, 180, 10), -- 10 meters tolerance
MDSYS.SDO_DIM_ELEMENT('Latitude', -90, 90, 10) -- 10 meters tolerance
8307 -- SRID for 'Longitude / Latitude (WGS 84)' coordinate system
-- CREATE THE SPATIAL INDEX --
-- Must be R-tree; quadtree not supported for geodetic data.
CREATE INDEX cola_spatial_idx_cs
ON cola_markets_cs(shape)
INDEXTYPE IS MDSYS.SPATIAL_INDEX;
--This search only return if distance is 10 000 km...
SELECT c.name
FROM cola_markets_cs c
WHERE
SDO_WITHIN_DISTANCE(c.shape,
SDO_GEOMETRY(2001, 8307, SDO_POINT_TYPE(150.0, 30.1, NULL), NULL, NULL),
'distance=10000 unit=KM')
= 'TRUE';
--According to this search the distance to point 1 is 3331198,72256398,
--this should be zero..
SELECT
c.mkt_id, c.name, SDO_NN_DISTANCE(1) dist
FROM cola_markets_cs c
WHERE SDO_NN(c.shape,
sdo_geometry(2001, 8307,sdo_point_type(150.0, 30.1, NULL), NULL, NULL),
'sdo_num_res=2', 1) = 'TRUE' ORDER BY dist

Hi,
What version of Oracle are you using? I got this using your example:
SELECT
c.mkt_id, c.name, SDO_NN_DISTANCE(1) dist
FROM cola_markets_cs c
WHERE SDO_NN(c.shape,
sdo_geometry(2001, 8307,sdo_point_type(150.0, 30.1, NULL), NULL, NULL), 'sdo_num_res=2', 1) = 'TRUE'
ORDER BY dist ;
MKT_ID NAME DIST
2 Point 2 0
1 Point 1 11085.3285
This is 10.1.0.4
Also:
Can I also combine SDO_WITHIN_DISTANCE with SDO_NN_DISTANCE so I can ask for all points 10 km in distance from the reference point and I can also see the actual distance for each point?
No, but you can add a distance calculation:
SELECT
c.mkt_id, c.name, SDO_GEOM.SDO_DISTANCE(c.shape,sdo_geometry(2001, 8307,sdo_point_type(150.0, 30.1, NULL), NULL, NULL),1) dist
FROM cola_markets_cs c
WHERE SDO_WITHIN_DISTANCE(c.shape,
sdo_geometry(2001, 8307,sdo_point_type(150.0, 30.1, NULL), NULL, NULL), 'distance=10 unit=km') = 'TRUE'
ORDER BY dist ;

[OT] Geographical points and the distance between them

I need to solve problems on the topic of addresses and the distance between them. I understand that I first need to convert an address into a pair of longitude and latitude. After my reading so far, I can use a GIS DB to achieve this goal. That is a lot of work since the issue is only one aspect of the application development. Does anyone know any other approaches?
Thanks.

[http://lmgtfy.com/?q=java%2Bcode%2Bdistance%2Bbetween%2Blatitude%2Blongitude%2Bpoints]

Distance between objects

I am using the Circle Distance example.vi (attached) and I want to improve the code. Imagine a picture with 4 circles. So instead of measuring the distance between one circle and 2 circles (VI example), I want to measure the distance between 1 circle and 3 circles ( see the picture attached).
How I can do that?
Attachments:
Distance image.png ‏22 KB
Circle Distance Example.vi ‏72 KB

If you look through the code it uses the IMAQ point distances VI to "[Compute] the distance between consecutive pairs of points."
What you want to do is essentially the same thing, but with a different set of inputs.
You just need to repeat some of your points to get the full combination.
Instead of just which is what I think the given example does1-2 2-3 1-3 (if you decrease the accuracy so it catches you squares you'll obviously go higher than 3)
You could do 1-2,1-3,1-4 , 2-3, 2-4, 3-4
I don't have time right now to confirm this is how it works, or a way to do this however.

Measuring distance between points

Similar Messages

Maybe you are looking for