I'm having some trouble understanding the wikipedia's pseudocode for connected-components labeling using Two-Pass algorithm. Here's the pseudocode:
algorithm TwoPass(data) is
linked = []
labels = structure with dimensions of data, initialized with the value of Background
First pass
for row in data do
for column in row do
if data[row][column] is not Background then
neighbors = connected elements with the current element's value
if neighbors is empty then
linked[NextLabel] = set containing NextLabel
labels[row][column] = NextLabel
NextLabel += 1
else
Find the smallest label
L = neighbors labels
labels[row][column] = min(L)
for label in L do
linked[label] = union(linked[label], L)
Second pass
for row in data do
for column in row do
if data[row][column] is not Background then
labels[row][column] = find(labels[row][column])
return labels
My problem is with the line linked[NextLabel] = set containing NextLabel. It never initializes the NextLabel and still uses it. Also, what does it mean by "set containing NextLabel"? I'm really confused by this code.
Related
I am trying to detect when two images correspond to a chunk that matches the other image but there is no overlap.
That is, suppose we have the Lenna image:
Someone unknown to me has split it vertically in two and I must know if both pieces are connected or not (assume that they are independent images or that one is a piece of the other).
A:
B:
The positive part is that I know the order of the pieces, the negative part is that there may be other images and I must know which of them fit or not to join them.
My first idea has been to check if the MAE between the last row of A and the first row B is low.
def mae(a, b):
min_mae = 256
for i in range(-5, 5, 1):
a_s = np.roll(a, i, axis=1)
value_mae = np.mean(abs(a_s - b))
min_mae = min(min_mae, value_mae)
return min_mae
if mae(im_a[im_a.shape[0] - 1:im_a.shape[0], ...], im_b[0:1, ...]) < threshold:
# join images a and b
The problem is that it is a not very robust metric.
I have done the same using the horizontal derivative, as well as applying various smoothing filters, but I find myself in the same situation.
Is there a way to solve this problem?
Your method seems like a decent one. Even on visual inspection it looks reasonable:
Top (Bottom row expanded)
Bottom (Top row expanded)
Diff of the images:
It might even be more clear if you also check neighboring columns, but this already looks like the images are similar enough.
Code
import cv2
import numpy as np
# load images
top = cv2.imread("top.png");
bottom = cv2.imread("bottom.png");
# gray
tgray = cv2.cvtColor(top, cv2.COLOR_BGR2GRAY);
bgray = cv2.cvtColor(bottom, cv2.COLOR_BGR2GRAY);
# expand rows
texp = tgray;
bexp = bgray;
trow = np.zeros_like(texp);
brow = np.zeros_like(bexp);
trow[:] = texp[-1, :];
brow[:] = bexp[0, :];
trow = trow[:100, :];
brow = brow[:100, :];
# check absolute difference
ldiff = trow - brow;
rdiff = brow - trow;
diff = np.minimum(ldiff, rdiff);
# show
cv2.imshow("top", trow);
cv2.imshow("bottom", brow);
cv2.imshow("diff", diff);
cv2.waitKey(0);
# save
cv2.imwrite("top_out.png", trow);
cv2.imwrite("bottom_out.png", brow);
cv2.imwrite("diff_out.png", diff);
Can the following expression of numpy arrays be vectorized for speed-up?
k_lin1x = [2*k_lin[i]*k_lin[i+1]/(k_lin[i]+k_lin[i+1]) for i in range(len(k_lin)-1)]
Is it possible to vectorize this calculation in numpy?
x1 = k_lin
x2 = k_lin
s = len(k_lin)-1
np.roll(x2, -1) #do this do bring the column one position right
result1 = x2[:s]+x1[:s] #your divider. You add everything but the last element
result2 = x2[:s]*x1[:s] #your upper part
# in one line
result = 2*x2[:s]*x1[:s] / (x2[:s]+x1[:s])
You last column wont be added or taken into the calculations and you can do this by simply using np.roll to shift the columns. x2[0] = x1[1], x2[1] = x1[2].
This is barely a demo of how you should approach google numpy roll. Also instead of using s on x2 you can simply drop the last column since it's useless for the calculations.
I have a google sheet that I want to add a conditional formatting, but the merged cells kinda of get in the way.
In my spreadsheet, I have the columns B,C,D and E where i'm using a formula in column B for when columns C,D and E are "done" column B gets a green background. The problem is, for some rows, columns D and E are merged.
This is the formula I'm using in column B
=and(C:C="done",D:D="done")
My desired result is: When the columns D and E are not merged, column B only gets the green background if columns C,D and E are "done", or else, it stays blank.
When D and E are Merged : B only gets green background if C and DE are "done", or else, stays blank.
Thanks in advance!!
If E always has a non-empty value when D and E aren't merged, you can use that to check for when E is blank using ISBLANK(E:E):
=AND(C:C="done", D:D="done", OR(E:E="done", ISBLANK(E:E)))
Otherwise, there isn't a way to test if a cell is part of a merged range without using a custom function, which is far less efficient, but could technically work with the assistance of a new helper column (e.g. column Z):
/**
* #customfunction
*/
function ISMERGED(cellAddress) {
var cell = SpreadsheetApp.getActiveSheet().getRange(cellAddress);
return cell.isPartOfMerge();
}
Where column Z (or whatever you choose for your helper) has the following in every relevant row (assuming data starts in row 1):
=ISMERGED(ADDRESS(ROW(E1), COLUMN(E1)))
=ISMERGED(ADDRESS(ROW(E2), COLUMN(E2)))
=ISMERGED(ADDRESS(ROW(E3), COLUMN(E3)))
...and so on.
And then for formatting B, use:
=AND(C:C="done", D:D="done", OR(E:E="done", $Z:$Z))
WARNING: If you toggle between merging and unmerging, Z wont contain the correct values immediately, though, because custom functions only re-compute when input changes (and the address of the cell wont in this case).
Update
Here's how you could compute your helper for the whole column downward, by row, by just putting the value in the topmost cell (ex: =ISMERGED("E:E") in cell Z1):
/**
* #customfunction
*/
function ISMERGED(rangeAddress) {
var range = SpreadsheetApp.getActiveSheet().getRange(rangeAddress);
var numCols = range.getNumColumns();
var numRows = range.getNumRows();
var result = [];
for (var i=0; i < numRows; i++) {
var rowRange = range.offset(i, 0, 1, numCols);
result.push(rowRange.isPartOfMerge());
}
return result;
}
best you can get is:
=OR(AND(C:C="done", D:D="done", E:E="done"),
AND(C:C="done", D:D="done", ISBLANK(E:E)))
which will work unless 4th row:
but you could pre-populate the whole E column with =CHAR(1) and then:
How my sheet works
I'm making a spreadsheet to show how much parts I have. By using a dropdown, am I able to show that I created a product. With conditional formatting I am showing that having 0 items isn't an issue when the product is created. Created products with 0 items change from red to purple. Purple means it doesn't matter to have 0 items from this product.
My issue
My issue starts with my dropdown. If I merge cells, The value will go into the upperleft cell. This means other cells inside the merged cell are blank. This gives me a problem with conditional formatting.
My conditional formatting code example:
=if($D2=0;$E2="Created")
I have to change this code for every cell because of the condition combined with a dropdown. Having more than 250 rows would be inhumanly hard to do by hand.
My questions
Are there ways to give all cells of a merged cell the value of the combined cell in an efficient way?
Is there a better way to make my conditional formatting code applyable to merged cells?
This is my sheet
Product items collected sheet link (Shows the problem and solution!)
Product items collected sheet image (Version 1)
Product items collected sheet image (Version 2)
At the heart of this question is the operation of merged cells. When a cell is merged, say over several rows, only the cell at the top left of the merged cell can contain data, respond to conditional formatting, and so on. In a manner of speaking the other cells cease to exist and values CANNOT be assign to them.
The questioner asks:
Q: Are there ways to give all cells of a merged cell the value of the combined cell in an efficient way?
A: No. Not just in an "efficient" way; it's just not possible.
Q: Is there a better way to make my conditional formatting code applicable to merged cells?
A: No and yes ;)
No. In so far as a merged cell is concerned, everything is driven by the value in the top cell of the merged range. There are no other options for the "rest" of the merged cell.
Yes. I'd create a "helper" cells in Column F as in this screenshot
The code to achieve this is dynamic - it will automatically adapt to adding more products, more items, etc.
The logic is fairly simple: Start in F2, test whether E2 has a value (that is, is it the top of the merged cell?). If yes, then assign the value of E2 to F2 AND put that value in a variable for the following cells. If no, the cell in Column E must be part of a merged cell, so assign the value for Column F to the variable that was saved earlier.
function so5270705902() {
// basic declarations
var ss = SpreadsheetApp.getActiveSpreadsheet();
// note this is going to work on the second sheet in the spreadsheet - this can be edited.
var sheet = ss.getSheets()[1];
// Column B contains no merged cells, and always contains data (it is the BOM for the Products).
// so we'll use it to established the last row of data.
var Bvals = sheet.getRange("B1:B").getValues();
var Blast = Bvals.filter(String).length;
// Row 1 is a header row, so data commences in Row 2 - this can be edited
var dataStart = 2;
// Logger.log("the last row in column D = "+Blast);// DEBUG
// set up to loop through the rows of Column F
var mergedcellvalue = "";
for (i = dataStart; i < (Blast + 1); i++) {
// set the range for the row
var range = sheet.getRange(i, 6);
//Logger.log("row#"+i+" = "+range.getA1Notation()); DEBUG
// get the value in column E
var ECell = range.offset(0, -1);
var ECellVal = ECell.getValue();
//Logger.log("offsetrange#"+i+" range value = "+ECellVal);
//Logger.log("Column E, row#"+i+", value = "+ECell.getA1Notation()+" range value = "+ECellVal);//DEBUG
// when a row is merged, on the top row contains any data
// so we'll evaluate to see whether there is any value in this row in Column E
if (ECell.isBlank()) {
//Logger.log("ECell is blank. We're in the middle of the Merged Cell"); ??DEBUG
// Set the value to the lastes value of "mergedcellvalue"
range.setValue(mergedcellvalue);
} else {
//Logger.log("ECell has a value. We're at the top of the merged cell");//DEBUG
// paste the ECellVal into this range
range.setValue(ECellVal);
// Update the "mergedcellvalue" variable so that it can be applied against lower cells of this merged cell
mergedcellvalue = ECellVal;
} // end of the if isblank
} // end of the loop through column F
}
UPDATE 22 October 2018
For development purposes, I used a small range of only 14 rows in Column E. However the questioner's data covers over 250 rows, so I expanded development testing to cover 336 rows (yeah, I know, but I was copy/pasting and I ended up with 336 and was too lazy to delete any rows. OK?). I found that the code took over 81 seconds to process. Not good.
The primary reason (about 80 seconds worth) for the long processing time is that there is a getValue statement within the loop - var ECellVal = ECell.getValue();. This costs about 0.2 seconds per instance. Including getValue in a loop is a classic performance mistake. My bad. So I modified the code to get the values of Column E BEFORE the loop
var Evals = sheet.getRange("e2:E").getValues();.
I was surprised when the execution time stayed around the same mark. The reason was that the isBlank evaluation - if (ECell.isBlank()) { which previously took no time at all, was now consuming #0.2 second per instance. Not good++. So after searching Stack Overflow, I modified this line as follows:
if (!Evals[(i-dataStart)][0]) {.
Including setValues in a loop is also asking for trouble. An option would have been to write the values to an array and then, after the loop, update the Column E values with the array. However in this case, the execution time doesn't seem to have suffered and I'm leaving the setValues inside the loop.
With these two changes, total execution time is now 1.158 seconds. That's a percentage reduction of , um, a LOT.
function so5270705903() {
// basic declarations
var ss = SpreadsheetApp.getActiveSpreadsheet();
// note this is going to work on the second sheet in the spreadsheet - this can be edited.
var sheet = ss.getSheets()[2];
// Column B contains no merged cells, and always contains data (it is the BOM for the Products).
// so we'll use it to established the last row of data.
var Bvals = sheet.getRange("B1:B").getValues();
var Blast = Bvals.filter(String).length;
// Row 1 is a header row, so data commences in Row 2 - this can be edited
var dataStart = 2;
// Logger.log("the last row in column D = "+Blast);// DEBUG
// set up to loop through the rows of Column F
var mergedcellvalue = "";
// get the values for Column E BEFORE the loop
var Evals = sheet.getRange("e2:E").getValues();
for (i = dataStart; i < (Blast + 1); i++) {
// set the range for the row
var range = sheet.getRange(i, 6);
//Logger.log("row#"+i+" = "+range.getA1Notation()); DEBUG
// get the value in column E
var ECell = range.offset(0, -1);
var ECellVal = Evals[(i - dataStart)][0];
//Logger.log("Column E, row#"+i+", value = "+ECell.getA1Notation()+" range value = "+ECellVal);//DEBU
// when a row is merged, on the top row contains any data
// so we'll evaluate to see whether there is any value in this row in Column E
// instead is isblank, which was talking 0.2 seconds to evaluate, this if is more simple
if (!Evals[(i - dataStart)][0]) {
//Logger.log("ECell is blank. We're in the middle of the Merged Cell"); //DEBUG
// Set the value to the lastes value of "mergedcellvalue"
range.setValue(mergedcellvalue);
} else {
//Logger.log("ECell has a value. We're at the top of the merged cell");//DEBUG
// paste the ECellVal into this range
range.setValue(ECellVal);
// Update the "mergedcellvalue" variable so that it can be applied against lower cells of this merged cell
mergedcellvalue = ECellVal;
} // end of the if isblank
} // end of the loop through column F
}
UPDATE 3 March 2019
The questioner made his final changes to the code. This code is the final solution.
function reloadCreatedCells() {
// Basic declarations.
var ss = SpreadsheetApp.getActiveSpreadsheet();
// Note this is going to work on the second sheet in the spreadsheet - this can be edited.
var sheet = ss.getSheets()[1];
// Column B contains no merged cells, and always contains data (it is the BOM for the Products).
// so we'll use it to established the last row of data.
var D_vals = sheet.getRange("D1:D").getValues();
var D_last = D_vals.filter(String).length;
// First row with data.
var dataStart = 2;
// Set up to loop through the rows of Column H - K.
var mergedcellvalue = "";
// Get the values for Column H - K BEFORE the loop.
var H_K_vals = sheet.getRange("H2:K").getValues();
// How many people we have.
var people = 4;
// The first vertical row.
var rowStart = 12;
// Horizontal rows.
for (var h = 0; h < people; h++) {
// Vertical rows.
for (var v = dataStart; v < D_last; v++) {
// Set the range for the row.
var range = sheet.getRange(v, rowStart + h);
// Logger.log(range.getA1Notation()); //DEBUG
// Get the value in column H - K.
var H_K_Cell = range.offset(0, -people);
// Adding Created and not created values inside L - O.
var H_K_CellVal = H_K_vals[(v - dataStart)][h];
// Logger.log(H_K_Cell.getA1Notation() + ': ' + H_K_CellVal); //DEBUG
// When a row is merged, the value is only inside the top row.
// Therefore, you need to check if the value is empty or not.
// If the value is empty. Place the top value of the merged cell inside the empty cell.
if (!H_K_vals[(v - dataStart)][h]) {
// Logger.log(H_K_Cell.getA1Notation() + ": is blank. We're below the top cell of the merged cell."); //DEBUG
// Set the value to the top cell of the merged cell with "mergedcellvalue".
range.setValue(mergedcellvalue);
} else {
// Logger.log(H_K_Cell.getA1Notation() + ": has a value. We're at the top of the merged cell."); //DEBUG
// Paste the H_K_CellVal into this range.
range.setValue(H_K_CellVal);
// Update the "mergedcellvalue" variable, so that it can be applied against lower cells of this merged cell.
mergedcellvalue = H_K_CellVal;
} // end of the if isblank.
} // End of the vertical row loop.
} // End of the horizontal row loop.
}
I am new to google scripts. I have a range where the cells in a column change. If the number of a cell from that column is above zero then the script has to send an email with the message being the content from a cell in the same row with the updated cell(that is, different column) Let's say If a value in column I changes then automatically it has to send an email with the message contained in the same row but column F for example. Please see below, I would appreciate some help.
function SendValue() {
var ss = SpreadsheetApp.getActive();
var sheet = ss.getSheetByName("Dividend");
var values = sheet.getRange("I2:I").getValues();
//var values1 = sheet.getRange("G2:G").getValues();
var results = [];
for(var i=0;i<values.length;i++){
if(values[i]>0){
results.push(sheet.getIndex([+i+2]+"G"));
// +2 because the loop start at zero and first line is the second one (I2)
}
}
MailApp.sendEmail('sxxxxxx#gmail.com', 'EX-DIVIDEND', results );
};
getValues() returns a 2 dimensional array, or more accurately an array of arrays, even if one of the dimensions is 1. values[i] is not a value but rather an array representing the i row of the range. You will need to acess that array to get at the cell at a particular column in that row.
replace if(values[i]>0) with if(values[i][0]>0)
I suspect that results.push(sheet.getIndex([+i+2]+"G")); should be results.push(sheet.getRange("G"+(i+2)).getValue()); as getIndex() doesn't take arguments and that looks like it is meant to be A1 notation.
However there is a better option than trying to construct A1Notation.
getRange() supports using numbers to refer to rows and columns. getRange(7, i+2) is a better choice.
https://developers.google.com/apps-script/reference/spreadsheet/range#getvalues
https://developers.google.com/apps-script/reference/spreadsheet/sheet#getRange(Integer,Integer)