1
00:00:00,000 --> 00:00:01,068
[No audio]

2
00:00:01,093 --> 00:00:04,590
Well, I hope you solve the exercise, and

3
00:00:04,590 --> 00:00:07,350
I believe it was not a difficult one. My

4
00:00:07,350 --> 00:00:09,810
purpose there was to get you used to

5
00:00:09,810 --> 00:00:13,650
with the OpenCV code, and also why not

6
00:00:13,680 --> 00:00:16,890
to practice a for loop. So, as you can

7
00:00:16,890 --> 00:00:19,740
imagine, this exercise will be solved by

8
00:00:19,740 --> 00:00:23,040
using a for loop. So, let me go through

9
00:00:23,040 --> 00:00:26,430
the code line by line quickly. So, we

10
00:00:26,430 --> 00:00:29,940
have to import OpenCV, so cv2 and the global

11
00:00:29,965 --> 00:00:33,020
library. If you can recall it

12
00:00:33,045 --> 00:00:36,390
what a globe does is that it finds a path names

13
00:00:36,420 --> 00:00:40,320
of some files given a certain pattern.

14
00:00:40,590 --> 00:00:43,410
So, in this case, for instance, I have

15
00:00:43,410 --> 00:00:48,030
this jpg files here, so, 1, 2, 3, 4, 5 images,

16
00:00:49,140 --> 00:00:52,920
and I said, Okay, create a list of file

17
00:00:52,920 --> 00:00:56,460
names that contain everything in the

18
00:00:56,460 --> 00:00:58,860
first part, and then jpg as

19
00:00:58,890 --> 00:01:02,388
extension. So that will create a list such as,

20
00:01:04,099 --> 00:01:07,115
say, like C and then the path

21
00:01:07,140 --> 00:01:10,500
here and then galaxy.jpg, and

22
00:01:10,500 --> 00:01:12,900
then the other image path and so on.

23
00:01:13,696 --> 00:01:15,930
You'll get the idea, and then what we need

24
00:01:15,930 --> 00:01:18,600
to do is iterate through this list for

25
00:01:18,600 --> 00:01:21,600
each path, image path in the list, in this

26
00:01:21,600 --> 00:01:25,170
list, we'll do these operations for each

27
00:01:25,170 --> 00:01:29,040
of the list items. So first, we will

28
00:01:29,040 --> 00:01:33,090
read that image path. Okay, that image

29
00:01:33,090 --> 00:01:35,430
file actually as a black and white

30
00:01:35,430 --> 00:01:39,030
image. So 0 is a flag is argument,

31
00:01:39,090 --> 00:01:41,760
which implies an image should write as

32
00:01:41,760 --> 00:01:43,890
black and white, in either grayscale

33
00:01:43,890 --> 00:01:46,050
actually, and then we create a variable

34
00:01:46,050 --> 00:01:48,120
where we will store the resized

35
00:01:48,120 --> 00:01:52,950
images, so the image is 100 by 100. So

36
00:01:52,950 --> 00:01:54,570
I'm passing the origin of the image here

37
00:01:54,600 --> 00:01:57,330
and their size, the new size that the

38
00:01:57,330 --> 00:01:59,700
image will get, and then we want to show the

39
00:01:59,700 --> 00:02:03,180
image just for demonstration, it's not

40
00:02:03,180 --> 00:02:05,400
really important, but it lets you check

41
00:02:05,400 --> 00:02:07,320
all the images that are being resized,

42
00:02:08,040 --> 00:02:10,320
and this is the name of the window, and

43
00:02:10,320 --> 00:02:14,520
then I pass here a waitKey method, and

44
00:02:14,520 --> 00:02:17,670
500 means 500 milliseconds. So each

45
00:02:17,670 --> 00:02:20,220
image will show and it will wait for

46
00:02:20,250 --> 00:02:22,920
half a second. So 500 milliseconds, and

47
00:02:22,920 --> 00:02:24,600
then after this half a second, Python

48
00:02:24,600 --> 00:02:26,820
will go to the next line and then to the

49
00:02:26,820 --> 00:02:29,190
next, and then it will go to the next item

50
00:02:29,190 --> 00:02:31,800
of the list and so on. So waitKey 500

51
00:02:31,800 --> 00:02:34,421
milliseconds, and then we have destroyAllWindows

52
00:02:34,445 --> 00:02:37,560
after this time passes, and then

53
00:02:37,560 --> 00:02:41,400
we write the resized_image. So re was

54
00:02:41,400 --> 00:02:44,610
a variable that holds the image object,

55
00:02:44,610 --> 00:02:47,220
the resized_image object, and this here,

56
00:02:47,670 --> 00:02:53,100
all this is a new name of the file. So

57
00:02:53,610 --> 00:02:55,860
what we get here is so we will have

58
00:02:55,860 --> 00:02:58,590
resized in the beginning of the image

59
00:02:58,590 --> 00:03:00,870
name and then just after that we

60
00:03:00,870 --> 00:03:03,570
will have a name of the original image.

61
00:03:04,740 --> 00:03:07,380
So for instance, we would have, for

62
00:03:07,380 --> 00:03:11,821
galaxy, we would have resized galaxy.jpg.

63
00:03:11,845 --> 00:03:15,255
So image here would be galaxy.jpg,

64
00:03:16,535 --> 00:03:18,930
which is this one here, and

65
00:03:18,930 --> 00:03:21,090
yeah, because the script is inside this

66
00:03:21,090 --> 00:03:24,120
folder, this list actually, the image

67
00:03:24,120 --> 00:03:25,980
of the list, would look something like

68
00:03:27,852 --> 00:03:34,140
galaxy.jpg, and then kangaroo,

69
00:03:34,670 --> 00:03:37,390
you know, australia.jpg, and so on. So

70
00:03:37,414 --> 00:03:40,052
[No audio]

71
00:03:40,089 --> 00:03:42,822
this gets the relative paths of the files.

72
00:03:42,870 --> 00:03:45,210
We're not getting the full path. Okay, I

73
00:03:45,210 --> 00:03:48,150
hope that is clear. So resized_galaxy,

74
00:03:48,150 --> 00:03:50,610
we could add underscore so that

75
00:03:50,610 --> 00:03:54,510
we discriminate the resized word from the

76
00:03:54,600 --> 00:04:01,020
other image name. Great, let me execute this

77
00:04:01,457 --> 00:04:02,697
python script.

78
00:04:02,721 --> 00:04:04,932
[No audio]

79
00:04:04,957 --> 00:04:06,990
So half a second, half a

80
00:04:06,990 --> 00:04:11,160
second, and yeah, its done, and let's check the

81
00:04:11,160 --> 00:04:14,670
images. So from here up are the original

82
00:04:14,670 --> 00:04:16,770
images and these are the image products.

83
00:04:17,010 --> 00:04:19,290
So you can see here that all these are

84
00:04:19,290 --> 00:04:22,685
100 by 100, and they are in a grayscale.

85
00:04:22,709 --> 00:04:24,709
[No audio]

86
00:04:24,734 --> 00:04:27,690
Okay, that's it. See you in the next lecture.

87
00:04:27,714 --> 00:04:29,714
[No audio]