Push.lua is great resolution handling library but some of its functions are limited in windows.
Especially when it comes to changing color.
And that is what I am having the problem with.
Whenever I give a specific color, it shows white.
Could anyone please tell me if there is any other resolution handling library for love2d on windows or if there is anyway I can get push.lua to work with changing some of its code.
RGB values must be a decimal number between 0 and 1 in the current version of love2d. By calling love.graphics.color(127, 45, 22) this will actually default to love.graphics.color(1,1,1) (which is the colour white) because 1 is the maximum value for R, G and B. If you want the RGB values of 127, 45 and 22 you should enter love.graphics.color(127/255, 45/255, 22/255) which is equivalent to love.graphics.color(0.498, 0.176, 0.086)
Related
I am trying to find the numbers in images like the one attached. I found that it helps when I run tesseract with the following parameters:
numbers = pytesseract.image_to_string(maskArray,config ='--oem 3 --psm 11 -c tessedit_char_whitelist=01234567890')
since I am looking for just numbers and they could be anywhere on the page. I found that tesseract can find 51416,51230, and 44133 but not 51523. At first I thought it may be a question of that dotted line behind the number, so I replaced that column with white. That did not seem to help. This image is already grayscale btw. One thing that seems to work is if I crop the image so that there is nothing left of that number. But in order to do that, I need to know where that number is. Any ideas what could be causing tesseract to randomly miss this number but not the other ones? or how to increase the accuracy?
Other things Ive tried:
rescaling up and down, inverting the image, blurring/unblurring, morphology/thresh,bilateral filter,erode , psm 11, 12, and default. Some combinations will increase the accuracy, but none seem to give me 100% which I would expect from the simplicity of the problem (currently maxing out at around 87%). Also, is there a way that I can increase the number of total outputs (including false positives)?
This is a bit of a long-shot. I really don't know where to ask this question.
I've been trying out CodeConnection + MakeCode with Minecraft and I haven't been able to figure out if there is correct way to place half-slabs at 0.5 step y axes increments.
I tried using a line between 2 points, but it left gaps between each slab.
If I try moving up 0.5, then it rounds it up to 1, and again leaves gaps.
It appears that all of the builder functions seem operate at a resolution of 1 block. However in-game I can obviously place slabs in 0.5 block increments to make stairs etc.
Blocks only exist at integer coordinates. Half slabs that exist in the top half of their space are still at a full integer coordinate. They just have a BlockState value of bottom=top (or top_slot_bit=true on Bedrock, represented by the integer value 8 as a bitflag, eg: 0b1... where the . bits are the integer representation of what type of slab (wood, stone, quartz...)).
What you're looking for is this widget, under Blocks:
You can set the block and then an integer representation of the desired data value (see the wiki on data values) in the numerical slot. This widget can then be dragged into the (block) portion of any block widget:
You'll probably have to some variable fiddling to get the data value to swap back and forth as you need it to, but that should solve the hurdle you've been facing.
I have failed after several attempts, and massive search attempts (which I may have just failed at), to determine how to move the text in the spreader title to accommodate use of the spreaderStartAngle and spreaderSliceAngle settings, which allow for changing the wheel to smaller sizes (180 or 90 degree versions in my case).
The demo on the site shows the text moving away from the center (starting as a 360 degree circle) down to a quadrant. But in my own use, the re-positioning does not occur automatically, and I see no additional attributes being applied that would cause this.
I have dived into more detail and found that there is apparently something in the Rafael library that may being modified (when looking at the source on the demo page), but it is beyond my understanding on how to approach this via the existing methods.
Anything I have tried makes no change to the text location, which is always centered in what would have been a full circle, regardless of the actual spreader's start and end.
You can position spreader's title via the spreaderPathCustom property.
wheel.spreaderPathCustom = spreaderPath().PieSpreaderCustomization();
wheel.spreaderPathCustom.titleRadiusPercent = 0.58;
wheel.spreaderPathCustom.titleSliceAnglePercent = 0.45;
PieSpreaderCustomization belongs to PieSpreader. The above values of titleRadiusPercent and titleSliceAnglePercent are suitable for 90 degree.
im working with SetConsoleTextAttribute API (Delphi to be specific), well anyway i cannot find a list of available colors anywhere?? can anyone help me out
There a 4 Bits used for the foreground color
FOREGROUND_BLUE; //1
FOREGROUND_GREEN; //2
FOREGROUND_RED; //4
FOREGROUND_INTENSITY; //8
which will give 16 possible colors. (0-15)
The next 4 bits are used for background color with the same scheme
See this. Colors are formed by ORing different constants. For instance:
An application can combine the foreground and background constants to
achieve different colors. For example, the following combination
results in bright cyan text on a blue background.
FOREGROUND_BLUE | FOREGROUND_GREEN | FOREGROUND_INTENSITY |
BACKGROUND_BLUE
For more Turbo/Borland Pascal CRTish implementations of such, see this. The colors available in the Windows console via SetConsoleTextAttribute work out roughly to be the same ones as the CRT unit, so any more relevant details can be found there.
Little description:
I get an image from my scanner using Capture Core Framework and background color is gray. So I want to do white background.
I'm looking for best way to remove this color. I looked at Image Core Filters. And as I understood I should use those ones, but I couldn't find included/ready filterfor this task. So, Do I have to write it myself?
It means to remove color I should write my own filter and apply it to my images? Right? Thank you.
UPD: I develop for Mac os
Why not just simply change the brightness of your image? If this gray background color value is fixed then you can increase the brightness a little bit, as a preprocess step, until you're satisfied.
A filter isn't quite the right tool for the job here. Filters apply a colour-shifting algorithm, pixel by pixel, to all of the pixels in an image.
This would be appropriate in your case if there was something completely unique about the pixels that you want to turn to white. For example, if the background you wish to eliminate was in a very narrow colour range that did not occur in any other part of the image. This is the technique used in greenscreen/bluescreen filming, which only works if that green or blue colour does not occur anywhere in the image regions you want to keep.
But - as you say in one of your comments - you cannot do this as that grey level you want to get rid of will not be unique to the background, so any filter you apply that would pick out those pixes may also affect pixels in the scanned subjectmatter.
What you really need is a way to select a region of interest and apply a filter to that region alone. You could use openCV for this. In fact it has a function that can achieve your result in one go:
http://docs.opencv.org/modules/imgproc/doc/miscellaneous_transformations.html?highlight=floodfill#int
floodFill(InputOutputArray image, //image to process
Point seedPoint, //starting pixel
Scalar newVal, //New value of the repainted domain pixels
Rect* rect, //optional output param (you won't need it)
Scalar loDiff, //max lower brightness/colour diff to select
Scalar upDiff, //max upper brightness/colour diff to select
int flags) //you want FLOODFILL_FIXED_RANGE
This function starts from a seedPoint, which should be any pixel that you can guarantee will be a part of the background grey you want to eliminate. (0,0) might work for you. It then interrogates neighbouring pixels, including them in the ROI array if they are sufficiently similar. The resulting array is a connected region. If your background grey uniformly falls between loDiff and upDiff - and your subject scan has a defined edge which does NOT fall into this range, you will get your result - selection and remapping of all background pixels to newVal (white).