File size: 2,729 Bytes
17e2002 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 | 1
00:00:01,030 --> 00:00:07,000
So let's quickly discuss sharpening sharpening as you imagine is the opposite of blurring and it actually
2
00:00:07,000 --> 00:00:12,310
strengthens strengthens and emphasizes the edges image as you can see in this example of sharpening
3
00:00:12,310 --> 00:00:12,740
here.
4
00:00:13,000 --> 00:00:15,040
The edges here are normal.
5
00:00:15,070 --> 00:00:20,050
What you would see with your own eyes however in this image all the edges here are much more pronounced
6
00:00:20,220 --> 00:00:22,110
physical horizontal.
7
00:00:22,240 --> 00:00:24,670
Are all the rooftops everywhere.
8
00:00:24,670 --> 00:00:26,170
So Templeman sharpening.
9
00:00:26,170 --> 00:00:32,940
We actually have to change our kernel and actually use the CV to filter to the function.
10
00:00:33,030 --> 00:00:35,850
So kernel for sharpening actually looks quite different here.
11
00:00:35,920 --> 00:00:42,730
However as you can tell it still seems to one that we we normalize our image other ways if it didn't
12
00:00:42,730 --> 00:00:48,410
normalize to one your image would be brighter or darker respectively.
13
00:00:48,490 --> 00:00:51,320
So let's run this simple sharpening example in our code.
14
00:00:51,550 --> 00:00:57,400
So again we look at image and then we create or sharpening Clennell says he saw before we have minus
15
00:00:57,400 --> 00:01:02,660
ones and all rows here and columns except indeed directions where we have nine.
16
00:01:02,700 --> 00:01:07,270
So if you would have some other elements in this matrix you'd actually get one which is exactly what
17
00:01:07,270 --> 00:01:07,860
we wanted.
18
00:01:07,960 --> 00:01:16,130
Which means it's normalized and to run or implement a sharpening function we use C-v to fill the 2d
19
00:01:16,760 --> 00:01:18,030
implemented shopping effect.
20
00:01:18,040 --> 00:01:23,400
So we take the shopping kernel and the input image we run it and we get a shop and image here.
21
00:01:23,440 --> 00:01:24,290
So let's take a look
22
00:01:29,450 --> 00:01:30,350
and there we go.
23
00:01:30,350 --> 00:01:34,330
This is our original image and this is a much sharper image.
24
00:01:34,340 --> 00:01:40,050
And as you can see just like the images we saw on the slide all the edges are much more pronounced.
25
00:01:40,190 --> 00:01:42,950
So it looks a bit artificial but you can play around with it couldn't.
26
00:01:43,040 --> 00:01:48,350
Metrics and trying it when shopping matrics They always to actually get a much nicer looking sharpening
27
00:01:48,440 --> 00:01:49,600
sharpened image here.
|