Metadata-Version: 2.4
Name: eckLEDurg
Version: 0.0.1
Summary: Computer Vision code for "seeing" LEDs with a camera automatically.
Author-email: Jakob Walker <Jakob.Walker@kratosdefense.com>
Project-URL: Homepage, https://code.devops.kratosdefense.com/projects/HWTST/repos/led-visual-detection/
Classifier: Programming Language :: Python :: 3
Requires-Python: >=3.7
Description-Content-Type: text/markdown
Requires-Dist: opencv-python>=4.11.0
Requires-Dist: opencv-contrib-python>=4.11.0

# eckLEDurg
This is a class that allows the user to point a camera at LEDs and determine how many are on

This class implements openCV to isolate and determines blobs of colors specified by the user

## The name
Remember that one book you had to read in high school. You know the one, the one where that one guy throws a ton of parties
and stares at green lights? Yeah, the name is a reference to that.

## How it works
The user specifies the Hue Saturation Value (HSV) model to look for a color and creates a mask of camera based on that color

The mask isolates the color specified and creates a black and white image where the white represented the desired color

A blob detector is then applied to the black and white image to detect where the white objects in the frame are

If these blobs are on screen in the same location for the amount of time specified by the user, their location in frame is returned

## An Explanation of Hue Saturation and Value

A full description of HSV can be found [here](https://en.wikipedia.org/wiki/HSL_and_HSV).

HSV represents all colors in a cylindrical format, where Hue represents the 360 degrees of the circle, Saturation represents
the distance away from the center of the cylinder, and Value represents the height. 

Hue: Represented by a range from 0 to 360. 0 represents red, 120 represents green, 240 represents blue, and 360 loops back
to represent red. Starting from red, the order of primary and secondary colors is red, orange, yellow, green, blue, purple, pink, red.

Saturation: Represented by a range of 0 to 100 and is used to represent how intense a color is. A Saturation of 0 is entirely
dull gray. A Saturation of 100 is entirely the color represented by Hue. For example, a Hue of 0 and a Saturation of 100 
represents a pure red color.

Value: Represented by a range of 0 to 100 and is used to represent how bright a color is. A Value of 0 represents black 
and a Value of 100 represents the brightest version of the color specified by Hue and Saturation. For example, a Hue of 0,
Saturation of 100 and Value of 100 represents the purest, brightest red possible using this model.

## Creating an LED Checker Instance

The following parameters are all necessary to instantiate the class:

`json_file` : The name of the JSON file that will store the configurations for the instance of an LED Checker. 

**NOTE:** This file does not have to exist, as it will be populated using `create_json_config`

`wait_time` : how long the program should look for LEDs in frame. range (0+)

`error_percent` : how accurate the LED identification should be. range (0-1)

`motion_tolerance` : the number of pixels off the center of a blob can be and still be counted as the same as another blob. range (0+)

For example, if one blob is found at (10,10) in one frame and then found in (10,11) the following frame, a `motion_tolerance` 
greater than or equal to 1 will count those blobs as the same blob

The following parameters are optional parameters:

`cv_capture` : The capture device used for the instance of an LED Checker 

See [Common Errors](#common-errors) for more details

Example instantiations:
    
    led_checker = LedCheck("example.json",5,1,10)
    led_checker_w_cv_capture = LedCheck("example2.json",3,12,4,cv.CAP_DSHOW)

## Class functions

### create_json_config

This function allows the user to see what the computer sees and manipulate it to isolate what the user specifies

When the function is called, it scans the device to find all cameras. If more than 1 camera exists, it presents a
list of camera indices for the user to choose from.

An explanation of camera indices can be found [here](#json-file)

After the camera has been chosen, the user will be presented with 2 windows: `Sharp Blobs` and `Object Detection`.

`Sharp Blobs`: A sharp black and white image that represents a manipulated frame to clearly create blobs. This frame will have
a lot of noise, this is OK. Changing hue will specify the color and changing saturation and value will adjust the amount of
noise seen in the frame.

Hue should have a range of about 60 units to fully encapsulate the color the user is looking for.  For example, if the 
user is looking to find a hue value for green, and the lower bound of hue is 80, the upper bound should  be around 140.

Saturation and Value ranges can vary, as long as the mask is what the user expects, the range can be as large or as small
as necessary.

This [site](https://colorpicker.me) has been provided to help visualize HSV values

**NOTE:** This website has accurate Hue values for colors but saturation and value are variable, please only use as a rough guide

`Object Detection`: A color image with red circles representing where the computer is seeing blobs in the image. Changing the area
will affect how large or small the white blob needs to be in order to be counted as a blob.

### check_led_status

This function checks if there is an LED matching the values within the JSON file specified during instantiation.

This function will display nothing to the user while it is running.

This function will return a list of (x,y) coordinates representing blobs that match all parameters given by the user.

## Example Execution

    green_led_check = LedCheck("green_led.json", 5, .95, 7)
    orange_led_check  = LedCheck("orange_led.json", 5, .95, 7)

    green_led_check.create_json_config()
    orange_led_check.create_json_config()

    green_results = green_led_check.check_led_status()
    orange_results = orange_led_check.check_led_status()

    print(f"Green: {green_results}, Orange: {orange_results}")

In this example, 2 LedCheck instances are created, `green_led_check` and `orange_led_check`

Both instances then call `create_json_config`, which brings up the windows that are used to configure what colors the LED
Check is looking for. Once these windows are closed, the configurations are saved to the JSON files specified in the constructor.

After the instances are configured to look for specific colors, `check_led_status` is called, which looks for the color
specified in the JSON file for 5 seconds (the specified amount of time). These points have a movement tolerance of 7 pixels
(the specifed amount of movement allowed) After that time, it validates which points are  there 95% of the time 
(the error percent specified).

Finally, the lists of (x,y) coordinates for both `green_led_check` and `orange_led_check` is returned to the user.

## JSON File
All configurations for the mask and blob detection are stored in a JSON file

`camera_index` : The index representing what camera to use.

    0 represents the built in camera on the computer
    1 represents an external camera plugged into the computer

**Note:** If there is no camera built into the device, an external camera will be represented by a 0.

`low_h` : The lower bound of the hue parameter to use in the mask (range 0:180)

`low_s` : The lower bound of the saturation parameter to use in the mask (range 0:180)

`low_v` : The lower bound of the value parameter to use in the mask (range 0:255)

`high_h` : The upper bound of the hue parameter to use in the mask (range 0:255)

`high_s` : The upper bound of the saturation parameter to use in the mask (range 0:255)

`high_v` : The upper bound of the value parameter to use in the mask (range 0:255)

`min_area` : The minimum size a blob has to be in order to be detected (range 0:5000)

## Common Errors

Depending on what camera and operating system the user has, the capture device may be different that the default.

Please use [this link](https://docs.opencv.org/3.4/d4/d15/group__videoio__flags__base.html) to determine what capture device to use

Once the capture device is determined, please use the built-in function `set_cv_capture` to change the capture device

**NOTE:** please format the parameter to this function as follows: `cv.CAPTUREDEVICE`

For example, if my capture device is `CAP_DSHOW`, them my function call would look like the following

    led_checker.set_cv_capture(cv.CAP_DSHOW)
