How the solutions are created:

Implementation 1

• Split 50% Click here NextPage
• This method is splitting each RGB Pixel and HSV Pixel on each video's frame by 50%.

• How to get HSV Pixel? In order to get HSV Pixel is converting each RGB Pixel to HSV(Hue Saturation Value) Pixel and each RGB Pixel must be split before convert it to HSV.

• Splitting RGB Pixel on each video's frame by 50% can be done by comparing each RGB Pixel to its Max Value means Red compared to its Max Value as well as Green and blue. Splitting HSV Pixel on each video's frame by 50% used the same splitting concept with RGB Pixel splitting concept which is comparing each HSV Pixel to its Max Value. Max Value of RGB and Max Value of HSV are different, Max value of RGB Pixel is 255 (Red: 255; Green:255; Blue:255), Max value of HSV Pixel; Hue is 360, Saturation and Value is 100.

• After we split each RGB Pixel and HSV Pixel, each split HSV Pixel must be converted back to RGB before we change / modify the RGB Pixel of each video's frame with the split RGB Pixel.

• Result:

Split 50%

                    

• User Selected Percentile Click here
• This method is splitting the RGB and HSV Pixels by 50% which is same concept with Split 50%.

• How does the splitting works? In this method, splitting can be done by comparing each RGB Pixel / HSV Pixel to the Percentile Slider and new value will be assigned to each RGB Pixel / HSV Pixel according to each max value of RGB Pixel / HSV Pixel.Max Value of RGB and Max Value of HSV are different, Max value of RGB Pixel is 255 (Red: 255; Green:255; Blue:255), Max value of HSV Pixel; Hue is 360, Saturation and Value is 100.

• The expected result depends on the User Selected Percentile or Slider of HSV and RGB. For HSV Percentile, how light and dark a color in each video’s frame will refer to Value Percentile, the dominance color of hue will refer to Saturation Percentile and the color will refer to Hue Percentile.
  

  
  
  

RGB Percentile

HSV Percentile

HSV Percentile With 0 Value

Implementation 2

• Equalization Click here
• This method works by creating a cumulative histogram using Fibonacci logic.

• How the cumulative histogram works? The frequencies (the number of pixels) at each intensity value are accumulated, the higher pixel density is the greater steepness is. In this assignment equalization will be applied by N Frame which N will be 1,5 and 20. Equalize by 1 frame means each frame will be equalize by applying the cumulative histogram to it, equalize by 5 frames and 20 frames are using the same concept / method as well as equalization by 1 frame, in equalize by 5 frames the equalization will be apply on every 5 frames, equalization by 20 frames means the equalization will be apply on every 20 frames.

Implementation 3

• Edge Detection & Smoothing
• This method works by applying convolution matrix operation to apply edge detection & smoothing.

• How the convolution matrix operation works? convolution matrix operation can be done by multiplying each pixel of each video's frame by using particular matrix as the mask. In this method, i'm using 3 x 3 Matrix as a mask and convolution matrix operation do multiply each pixel of each video's frame to the correspond matrix value and the surrounded 8 pixel do the multiplication to the correspond matrix value as well and sum each multiplication result. The multiplication result will be divided by the divisor. Divisor is the sum of matrix value according to the matrix size.

• Matrix used for edge detection & Smoothing:

                    Edge Detection                        Smoothing
                    -1 -1 -1                                             1 2 1
                    -1  8 -1                                             2 4 2
                    -1 -1 -1                                             1 2 1

                    

                    By using 3 x 3 matrix above, the divisor of edge detection will be 16 and the divisor of
                    smoothing will be 0 but in my implementation if the divisor equals to 0, the divisor
                    value will be 1 and the multiplication result will be divided by 16.The new filtered pixel    
                    will be assigned to original pixels of each video's frame. 

Edge Detection

Smoothing

About us