Recently Updated
Hot or Not
Challenge
According to my friend Zuck, the first step on the path to great power is to rate the relative hotness of stuff… think Hot or Not.
(this is a scaled-down image, original was >70Mb)
Solution
Looking at the image more closely, we see it is made up of a series of subimages of either dogs or hotdogs.
Looks like we have to classify the subimages into hotdogs or regular dogs..
First we split the image up into all its subimages with imagemagick
1
$ convert -crop 224x224 +repage hotdogs/out%04d.jpg
Next, we can use Clarifai to do the image recognition to determine whether the subimages are dogs or hotdogs. Clarifai gives you 5000 free operations per month, but since we have a little over 7500 subimages, we needed two accounts to perform this analysis.
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
import os
import json
from PIL import Image
from clarifai.rest import ClarifaiApp
from clarifai.rest import Image as ClImage
app = ClarifaiApp(api_key='f7f15032b1a04f1fafc2092c63e50e9f')
model = app.models.get('general-v1.3')
# detect image contents for all subimages
pixels = []
for i in range(0,87*87):
image = ClImage(file_obj=open("hotdogs/out"+str(i).zfill(4)+".jpg", 'rb'))
response = model.predict([image])
hot = False
concepts = response['outputs'][0]['data']['concepts']
for concept in concepts:
if 'food' in concept['name']:
hot = True
pixels += [1 if hot == True else 0]
# make qr code image
(w,h)=(87,87)
outimg = Image.new( 'RGB', (w,h), "white")
pixels_out = outimg.load()
p = 0
for i in range(0,h):
for j in range(0,w):
print(pixels[p])
if pixels[p] == 1:
pixels_out[j,i]=(0,0,0)
else:
pixels_out[j,i]=(255,255,255)
p += 1
outimg = outimg.resize((5*w,5*h))
outimg.save("pixels_outimg2.png","png")
This outputs the following image:
This is clearly a QR code, it is just missing the corner anchors. We add these and clean the image up slightly:
Flag
1
IceCTF{h0td1gg1tyd0g}
Flag
IceCTF{h0td1gg1tyd0g}