To one of us, seekers of the truth. We realise th...

创建于：2026年4月11日

问题

report.pdf

To one of us, seekers of the truth.

We realise that you wanted to be left out of this after what happened last time, but we are in a dire need of your expertise. We recently lost contact with the Professor after receiving his urgent report. We think that some kind of clue can be hidden in the report, but we are not able to find it. We hope that, as you worked with the Professor for some time, you will be able to find it.

问题

ninja-nerds.zip

where are your little ninja-nerds?
find flag format: UMASS{flag}

Python

python
from PIL import Image
img=Image.open('/mnt/data/ninja/challenge.png').convert('RGB')
img.size


(640, 360)

Python

python
from PIL import Image
import numpy as np, os, textwrap, math
img=Image.open('/mnt/data/ninja/challenge.png').convert('RGB')
arr=np.array(img)
outdir='/mnt/data/ninja_bits'
os.makedirs(outdir, exist_ok=True)
for ch,name in enumerate('RGB'):
    for bit in range(8):
        plane=((arr[:,:,ch]>>bit)&1)*255
        Image.fromarray(plane.astype('uint8')).save(f'{outdir}/{name}_bit{bit}.png')
# combined lsb of all channels maybe
for bit in range(8):
    plane=((arr>>bit)&1).mean(axis=2)
    plane=(plane*255).astype('uint8')
    Image.fromarray(plane).save(f'{outdir}/avg_bit{bit}.png')
print('saved', outdir, len(os.listdir(outdir)))

Python

python
from PIL import Image
import numpy as np, itertools, re, math, os, string, collections
img=Image.open('/mnt/data/ninja/challenge.png').convert('RGB')
arr=np.array(img)
h,w,c=arr.shape
channels={'R':arr[:,:,0].flatten(),'G':arr[:,:,1].flatten(),'B':arr[:,:,2].flatten()}
orders=['RGB','RBG','GRB','GBR','BRG','BGR']
# gather bytes from selected bitplanes orders
def bits_to_bytes(bits):
    # truncate to multiple of 8
    n=len(bits)//8*8
    bits=np.array(bits[:n], dtype=np.uint8)
    by=np.packbits(bits)
    return bytes(by.tolist())
def ascii_spans(data, minlen=6):
    out=[]
    cur=[]
    start=0
    for i,b in enumerate(data

(0, [])

思考

undefined

Python

python
from PIL import Image
import numpy as np, itertools, re, math, collections, binascii, base64
img=Image.open('/mnt/data/ninja/challenge.png').convert('RGB')
arr=np.array(img)
h,w,_=arr.shape

# various sequences of source byte arrays
seqs={}
flat_channels={ch:arr[:,:,i].flatten() for i,ch in enumerate('RGB')}
# row-major interleaved RGB
for order in itertools.permutations('RGB',3):
    seqs['pix_'+''.join(order)] = np.stack([flat_channels[c] for c in order],axis=1).reshape(-1)
# row-major single channels
for c in 'RGB':
    seqs[c]=flat_channels[c]
# column-major
arrT=arr.transpose(1,0,2)
flatT={ch:arrT[:,:,i].flatten() for i,ch in enumerate('RGB')}
for order in itertools.permutations('RGB',3):
    seqs['col_'+''.join(order)] = np.stack([flatT[c] for c in order],axis=1).reshape(-1)
for c in 'RGB':
    seqs['col_'+c]=flatT[c]

def packbits_with_offset(bits, offset=0, reverse_each=False):
    if offset:
        bits=bits[offset:]
    n=len(bits)//8*8
    bits=bits[:n]
    if reverse_each:
        bits=bits.reshape(-1,8)[:,::-1].reshape(-1)
    return np.packbits(bits).tobytes()

flag_re = re.compile(rb'UMASS\{[^}]{0,200}\}')
results=[]
for name,seq in seqs.items():
    for bit in range(8):
        bits=((seq>>bit)&1).astype(np.uint8)
        for off in range(8):
            for rev in [False, True]:
                data=packbits_with_offset(bits, off, rev)
                m=flag_re.search(data)
                if m:
                    results.append((name,bit,off,rev,m.group()))
                # try strings with braces
                for pat in [rb'UMASS', rb'\{', rb'flag', rb'ctf', rb'FLAG']:
                    idx=data.find(pat)
                    if idx!=-1:
                        results.append((name,bit,off,rev,pat,idx,data[idx:idx+80]))
                        break
len(results), results[:20]


(451,
 [('pix_RGB',
   0,
   0,
   False,
   b'\\{',
   9977,
   b"\\{\xd4\xcfI \x00\x00\x00\x00\x01\x01\x0f\xf2\xd1\xd8\r\xb9\x12\x9bm\xb6\xdb\x17'\xebL\xce\x0c\x9em-\x99\x06BI\xb6\x93m\x80\xef\xb3\xdd\xb7\xff\xff\x92M\xa2\xd8N\xfb\xd3l\x86\xdf\x95\xc9\xb7q\xd79\xc3\xb68\t\x92=\xfb\xb4\xeeI&v\xfd\xa2\xe5\xb96\x92"),
  ('pix_RGB',
   0,
   0,
   True,
   b'\\{',
   2244,
   b'\\{ \xa6\x90\xb5\xe4\xff\x93v\xb2\xb5\xaf\x1a\xda\xed\xff\xff\xdb\xb6I\x12\xfe\x87\x1d?\x99O\x9c\xfc\t\x00 \x95\x80\xc8\x8d\x93\rB\xa4\xa1|Y\xda#\xb9m\x1f\xaeor\xdf\xdb\xc0\x83\x0bm\xdf\xb6\xf6\xff\xb7m\xdb\xd6\x87\x86\xe2\x95\\\xba\xff\xff\xff\xfa\x15I$\xad'),
  ('pix_RGB',
   0,
   1,
   False,
   b'\\{',
   83514,
   b"\\{lI;l\xdbm\xb9\t)%\xb6\xc0\x01\xff\xff\x80\x95\xa9u\xc3r6\xdbm\xb6\xdf\xe0l\x06\xc6\x7f\xb7U\xda\xdb?\x81\xff;p\x00\x1f\x16\xed\xa9%m\xdbm\xb1\x1c\x7f\xff\xe0+\xc9-\xb6I6\xff\xf8\xe2\x9c\x8a\xa6\xdbg\x89~\x04m\xb7'\x00LI?"),
  ('pix_RGB',
   0,
   1,
   True,
   b'\\{',
   45457,
   b"\\{\x9c\xb4}$,\xca\x07\xc0/p\x9eg\xb27\x01t\xdbC\x85\xf0?\xa1\xbf\xf0\xc3\xff\xc3\x84\xb5\xff\x88QLH\x88\xcaI\x9e\x99\xb0\xbc7\x16\xca\xc4\xd1\xac%\xe9\xd8\xf1\xbb\x0b\xd1!`H&n['\xbe(I\x92\x12\th\xdb\xe9\xbd\xcd\xff\x8f&\x85\xb3a"),
  ('pix_RGB',
   0,
   2,
   True,
   b'\\{',
   44570,
   b'\\{\xdc\xa4\xc5\x06\xb2\x98\xfb\xff\xdf\xb6m\xdb\xb6m\xdb\xfe?\x00\xfcX\x8f\xd7\xcf\x06n)Lv\x15\x99%\x08d\xf4\x99b\xe5N\xcf\x9dD?\xacf\x11x\xa6Q\xf7l"I\x15\xb6\x01\x03b\x03\x00I\xc7N\xb6\xdb.\xb6m\x9b\x8fm\xb1\x13\xe0\xa7\x001\xdf\x02'),
  ('pix_RGB',
   0,
   3,
   False,
   b'\\{',
   38166,
   b'\\{[\xff\xa5U\xfe\xdb\xff\xb7\xffY2I%\xb5$\x00@_\xef\\\x01\xc5\xa9$\x9a\xc8\x0btss=\x9ei\xbf\x85\xbd\x14\x9bm\x87F9\xfb\xd0f\xe5\xb6\xc8\xdd\xe5\x1f\xc2\xeb\xec\x9b\xc9s\x1b>=\x16lqPG\xf6\xe9\xf5\xa5\xdd\x99\x8fU\x1c^\xb3\xd1\xaa'),
  ('pix_RGB',
   0,
   4,
   False,
   b'\\{',
   27250,
   b"\\{\xff'\xdf\xefY{\xe8\xfd\xd4\tw&\xe3}\xcb\x9dE-\x80\xab\xf4\x95\x03i\t\xff\xfe\n\x92F\xce\x93\x00-\xb4u\xa9)\x9eY\xa0\x81$\x14\x99\x0e\\\xe0\xde\x1b\xf0\x82\x8c\x12\x92I$\x92&I6\x92\xb1f\xf8F\xab\xa4\xad\xb6\xcf\x10\xd8\x886\xdb\x7f\xe2"),
  ('pix_RGB',
   0,
   4,
   True,
   b'\\{',
   37802,
   b'\\{\xb9(Mx\x81\x88\xdc $\x1a\x8a\xbf\xf3\xe9\x84R\x95\xc6\x81\x90m\xdb0\xd2\xb8\xd0,1\xcb<\x04P\x80\x7f\xb6\x01\\b\xff\xaa|{/\xddY&\xd3<\x02W\xd6E\xeb0\xc9\xf0\x97\x06\xbc\xc4\xff\x9f\xa9jK\x92\xdbF\x92\x00\xd0\xb6m\xdb\xb6\x92\xf4\x1f'),
  ('pix_RGB',
   0,
   5,
   False,
   b'\\{',
   28978,
   b'\\{\xdf\x17\xe1l\xcfa\xa7$\x93la\xa9I\xb6\xd6\xdf\x89\x02\xda\x11\xc4\x92I$\x92I;\x1f\xf8\x00~\x00\x01$\x00\x16\xfe\x00\x1b\xecb"D\xf7\xff\x81\xb5\xb2ZI$\x93$\x9bZI$\xb9$\x95+U\xa5\x83\xad\xd8\xa1\xef\xb4\xdd\x8a\xee\x81y`\x86\xa1\xf2'),
  ('pix_RGB',
   0,
   6,
   True,
   b'\\{',
   22465,
   b'\\{\x03\xdeY\x0e\x0f(n/u\x95\x9d\xfc\xc0\xdb\x02\xc0+\xfdR\xfeu\xc2\xa4P\xddk\xdd\xb6m\xdb\xb6m\x93\xae0\xd0.\xab\x85$I\x92$I\x92\xe4\xb6m\x1bI\x92\xe46\xc1W\xc3\x83V\xeb\x9aG\xc6\x85\x97m\xdb6In\x1b\xdf\xbe\x9a=\xf3\xdb\xf5\xa3'),
  ('pix_RGB',
   0,
   7,
   True,
   b'\\{',
   37619,
   b'\\{\xcc\xd00\x91\x94\xd0&Ir\xdbH\x12\x00\xda\xb6m[I\x92\xfe\x1f\x80\xa0\xa2\x9fE\x0f3\xa2k\x00$I\x92\x00\x00@\xeem\xa3\x11\x0b\x97\xc8DUm{Il\x13\x00\xde\x1a\x85\xfe\xff\xff\xdb\x90\x1c\xb8`I*I\xc0\xb2m\xdbH\xb2m\xdb\xb6\xb5\xc9$'),
  ('pix_RGB',
   1,
   0,
   True,
   b'\\{',
   5531,
   b'\\{\xdc\xf56I\xf71\x13u/\x88\xd0\x98\x1f\x98\x13 I:\xa7\xe4\x9b\x91\xed\x8b\\m\xed\xff\xff\x12\t$I\xf2\xdbR\xdb\xb6\x94jI\x01\xa9o\xdb\xb6\r\xcd\xe6\x1e\xff\xdf\xb6I\x92\xb4t\x15\xc9\xc3]\xff\xfe\xbfm\x92\xda\xb6\xedO\x12\xd77G\xd2\x97\x10"'),
  ('pix_RGB',
   1,
   1,
   True,
   b'\\{',
   22908,
   b'\\{\xbf\x93\xc1\x14{B\xe2\xaf\xff\xc7\xaa*\x0e\x93\xc9\x8f\xb5\xeeD`\x1b\xe3\xa0\t\t^=\xfc\xaf\xd3f\x80\x12\xb4\x92\xdc\\dM >\x84\xa5K\x03l\xb4\xdem\x8b\xd86\x92\xb4\x93\xb6\xfd\x07u`\xbd9\xbcx\x91\xc0\xb6\xb6m\xdb\x8d\x9crn\xf4\xe2n\xdb'),
  ('pix_RGB',
   1,
   2,
   False,
   b'\\{',
   44986,
   b"\\{\xf0Y$\xe3Zq\xd3\xfd+\xc0\x9dVe:2\x1a\xa1\x93'/\xc3\xec\x00y,\x1cU\xb2 +\xbb]\xa7\xc1\xd6\x06\xc8\xa6\x01\xb6\xc0\x03\xff\xdbm\xa4\xe4\x92O\xd6]5\xfc\xb6\xdbI\xfd)$n\xd2\x1c\xa6\x9a\x9a70i\xe3\xcex\xff\xf2I$\x92I$"),
  ('pix_RGB',
   1,
   2,
   True,
   b'\\{',
   17070,
   b'\\{*Z_=\xcf\xfb\rmKB\x02u\xc5e\xe8BF`\xd6\x87\x03\x90\x00\x96\x1f\xb6\x01@\xdb\xb6m\x02)\xb3$\xed\xe2\xb6MX\x11\xdb\x1e\xb0\xa3j[\x92u3\xc3\x96\\`V\xa3\xdc\xdf\x85\xf0t\xb6\xb4\x00I&\tI\x92\xe4\xdc\xc0\xf2x\xa7\x0c\xdc\xf6'),
  ('pix_RGB',
   1,
   3,
   False,
   b'\\{',
   41186,
   b"\\{\x9b\xbc\x00m\xbc\xb6\xdbW\xff\xa4r\xfb{\xc2\xf9\xc6\xa2\xe4N\xb9A<j+\xff\xdc\x147|\x00\x12\x06\xdbm\xb6\xdbm\xb3m\xfd\x81'T\x94\x92I%\xb6\xa4\x91\x9f\x0f\x04\x86r!H\xb4\xb1\xad4\x8aS\x87\xa2I\x00\x04\x8d\xb7\n\x01\xeb\xeam\xdb\x85\xa4"),
  ('pix_RGB',
   1,
   4,
   False,
   b'\\{',
   5654,
   b'\\{w9\x8d\x04\xb7Z+\xfbf\xc4\xe5$\x98D\x07\x1d\xd4b4\xfbI$\x9f\xff\x18\x81\x033\x93\xff\x00\x0b%C\x9b\x00\xdbbm\xb6\xd0\x00\x00\t$\x92@\x00$\x9d\xb6\xdbf\xdbr@\x00\x00\x0bmI-\xb6\xd9$\x93\xe3S\x97\xd0pc\xd7\xed\xb6\x00\t$'),
  ('pix_RGB',
   1,
   4,
   True,
   b'\\{',
   63332,
   b'\\{)\xf8S\xdaZ\x92$y\xbdlK\xf2\xc3\xbf\x91@\x92\xc4\x0f\x1e2Lv$\x92\xd4\xb6\x85$)\x10\x90$-IV\x1d\xc7M\x80$o\xdb\xb6\x8d\xe6V\x81\xa2n\xdbj\xe1\xcdK\x1cRm\xdb\xb6q\xacJZN\x02d\xc8]\xb6\rq2z\x00\xdb\xdc\xff'),
  ('pix_RGB',
   1,
   5,
   True,
   b'\\{',
   4805,
   b'\\{\x0b\xd2\xd3qV\x9a\x18\xb7%\x88\xe4\xd7D\xc6.\xf2\t.\xd1\x9c$I\xd2:\xf5\xd5\xd5G\xf5\xc8u\x92\xa4\xff\xff\xff\xc4\xb6\x81\xe4\x97$I[\x80i-H}\xdb\xb6mcw4\xf8\xff\xb6M\x92\xa4%\x17\xacR\xdc\xb6\xf6\xff\xff\x7f\xdb\xb6m\x7f\x92s\x82'),
  ('pix_RGB',
   1,
   6,
   True,
   b'\\{',
   34170,
   b'\\{\x1d\xc8\xb3"\x8e\xd5.Y!\x0b!E?@={v\x11\xcd>\xe9\xf3j\xfb\xa2 \x9e#\x89\xd3E\xd0\xafO\xcd\xf3+\x0f\xb1\xab\xdb\xb5\x9es&\n\xed\xff/\x0ev\x1e\x01\xa9\x7f\xe0\xd72\xf6/\x01$\xf9s\xfb\xe7?\x8f\xe6,\xa4\r\xfc\'%C \xe9')])

Python

python
import re, itertools, numpy as np, binascii
magic = {
    b'PK\x03\x04':'zip',
    b'\x89PNG\r\n\x1a\n':'png',
    b'%PDF':'pdf',
    b'GIF87a':'gif',
    b'GIF89a':'gif',
    b'\xff\xd8\xff':'jpg',
    b'7z\xbc\xaf\x27\x1c':'7z',
    b'Rar!\x1a\x07':'rar',
    b'OggS':'ogg',
    b'ID3':'mp3',
    b'<?xml':'xml',
    b'{':'brace',
    b'UMASS{':'flag'
}
hits=[]
for name,seq in seqs.items():
    for bit in range(8):
        bits=((seq>>bit)&1).astype(np.uint8)
        for off in range(8):
            for rev in [False,True]:
                data=packbits_with_offset(bits, off, rev)
                # search magic
                for m,typ

([('pix_RGB', 0, 0, False, 'brace', 1648),
  ('pix_RGB', 0, 0, True, 'brace', 1646),
  ('pix_RGB', 0, 1, False, 'brace', 690),
  ('pix_RGB', 0, 1, True, 'brace', 771),
  ('pix_RGB', 0, 2, False, 'brace', 37),
  ('pix_RGB', 0, 2, True, 'brace', 322),
  ('pix_RGB', 0, 3, False, 'brace', 1464),
  ('pix_RGB', 0, 3, True, 'brace', 9),
  ('pix_RGB', 0, 4, False, 'brace', 1404),
  ('pix_RGB', 0, 4, True, 'jpg', 8768),
  ('pix_RGB', 0, 4, True, 'brace', 347),
  ('pix_RGB', 0, 5, False, 'brace', 730),
  ('pix_RGB', 0, 5, True, 'brace', 732),
  ('pix_RGB', 0, 6, False, 'brace', 770),
  ('pix_RGB', 0, 6, True, 'brace', 2458),
  ('pix_RGB', 0, 7, False, 'brace', 12),
  ('pix_RGB', 0, 7, True, 'brace', 689),
  ('pix_RGB', 1, 0, False, 'brace', 374),
  ('pix_RGB', 1, 0, True, 'brace', 489),
  ('pix_RGB', 1, 1, False, 'brace', 690),
  ('pix_RGB', 1, 1, True, 'brace', 35),
  ('pix_RGB', 1, 2, False, 'brace', 323),
  ('pix_RGB', 1, 2, True, 'brace', 24),
  ('pix_RGB', 1, 3, False, 'brace', 34),
  ('pix_RGB', 1, 3, True, 'brace', 95),
  ('pix_RGB', 1, 4, False, 'brace', 2665),
  ('pix_RGB', 1, 4, True, 'brace', 131),
  ('pix_RGB', 1, 5, False, 'brace', 2177),
  ('pix_RGB', 1, 5, True, 'brace', 21),
  ('pix_RGB', 1, 6, False, 'brace', 484),
  ('pix_RGB', 1, 6, True, 'brace', 373),
  ('pix_RGB', 1, 7, False, 'brace', 96),
  ('pix_RGB', 1, 7, True, 'brace', 208),
  ('pix_RGB', 2, 0, False, 'brace', 243),
  ('pix_RGB', 2, 0, True, 'brace', 149),
  ('pix_RGB', 2, 1, False, 'brace', 805),
  ('pix_RGB', 2, 1, True, 'brace', 66),
  ('pix_RGB', 2, 2, False, 'brace', 743),
  ('pix_RGB', 2, 2, True, 'brace', 244),
  ('pix_RGB', 2, 3, False, 'brace', 66),
  ('pix_RGB', 2, 3, True, 'brace', 109),
  ('pix_RGB', 2, 4, False, 'brace', 247),
  ('pix_RGB', 2, 4, True, 'brace', 1181),
  ('pix_RGB', 2, 5, False, 'brace', 109),
  ('pix_RGB', 2, 5, True, 'brace', 14),
  ('pix_RGB', 2, 6, False, 'brace', 992),
  ('pix_RGB', 2, 6, True, 'brace', 233),
  ('pix_RGB', 2, 7, False, 'brace', 73),
  ('pix_RGB', 2, 7, True, 'brace', 804),
  ('pix_RGB', 3, 0, False, 'brace', 260),
  ('pix_RGB', 3, 0, True, 'brace', 1943),
  ('pix_RGB', 3, 1, False, 'brace', 49),
  ('pix_RGB', 3, 1, True, 'brace', 1924),
  ('pix_RGB', 3, 2, False, 'brace', 615),
  ('pix_RGB', 3, 2, True, 'brace', 1183),
  ('pix_RGB', 3, 3, False, 'brace', 155),
  ('pix_RGB', 3, 3, True, 'brace', 59),
  ('pix_RGB', 3, 4, False, 'brace', 223),
  ('pix_RGB', 3, 4, True, 'brace', 97),
  ('pix_RGB', 3, 5, False, 'brace', 2720),
  ('pix_RGB', 3, 5, True, 'brace', 2398),
  ('pix_RGB', 3, 6, False, 'brace', 97),
  ('pix_RGB', 3, 6, True, 'brace', 565),
  ('pix_RGB', 3, 7, False, 'jpg', 34857),
  ('pix_RGB', 3, 7, False, 'brace', 2027),
  ('pix_RGB', 3, 7, True, 'brace', 48),
  ('pix_RGB', 4, 0, False, 'brace', 486),
  ('pix_RGB', 4, 0, True, 'brace', 126),
  ('pix_RGB', 4, 1, False, 'brace', 37),
  ('pix_RGB', 4, 1, True, 'brace', 246),
  ('pix_RGB', 4, 2, False, 'brace', 1231),
  ('pix_RGB', 4, 2, True, 'brace', 1789),
  ('pix_RGB', 4, 3, False, 'brace', 246),
  ('pix_RGB', 4, 3, True, 'brace', 80),
  ('pix_RGB', 4, 4, False, 'brace', 150),
  ('pix_RGB', 4, 4, True, 'brace', 6),
  ('pix_RGB', 4, 5, False, 'brace', 80),
  ('pix_RGB', 4, 5, True, 'brace', 536),
  ('pix_RGB', 4, 6, False, 'brace', 6),
  ('pix_RGB', 4, 6, True, 'brace', 485),
  ('pix_RGB', 4, 7, False, 'brace', 296),
  ('pix_RGB', 4, 7, True, 'brace', 150),
  ('pix_RGB', 5, 0, False, 'brace', 168),
  ('pix_RGB', 5, 0, True, 'brace', 965),
  ('pix_RGB', 5, 1, False, 'brace', 3088),
  ('pix_RGB', 5, 1, True, 'brace', 296),
  ('pix_RGB', 5, 2, False, 'brace', 965),
  ('pix_RGB', 5, 2, True, 'brace', 2454),
  ('pix_RGB', 5, 3, False, 'brace', 408),
  ('pix_RGB', 5, 3, True, 'jpg', 36272),
  ('pix_RGB', 5, 3, True, 'brace', 11966),
  ('pix_RGB', 5, 4, False, 'brace', 6266),
  ('pix_RGB', 5, 4, True, 'brace', 3413),
  ('pix_RGB', 5, 5, False, 'brace', 12553),
  ('pix_RGB', 5, 5, True, 'brace', 1974),
  ('pix_RGB', 5, 6, False, 'jpg', 37711),
  ('pix_RGB', 5, 6, False, 'brace', 6293),
  ('pix_RGB', 5, 6, True, 'brace', 7442),
  ('pix_RGB', 5, 7, False, 'brace', 7706),
  ('pix_RGB', 5, 7, True, 'brace', 5065)],
 2289)

思考

undefined

思考

undefined

Python

python
from PIL import Image
import numpy as np, matplotlib.pyplot as plt, os, math
img=Image.open('/mnt/data/low4g_big.png').convert('L')
arr=np.array(img)
arr.min(), arr.max(), arr.mean(), arr.std()


(np.uint8(0),
 np.uint8(255),
 np.float64(127.59426649305556),
 np.float64(46.35677289611709))

Python

python
import numpy as np, cv2, os, math
arr=np.array(Image.open('/mnt/data/low4g_big.png').convert('L'))
# try adaptive threshold after blur
for blur in [0,1,2,3]:
    a=arr.copy()
    if blur:
        a=cv2.GaussianBlur(a,(0,0),blur)
    for tname,thr in [('dark',60),('bright',200)]:
        if tname=='dark':
            bw=(a<thr).astype(np.uint8)*255
        else:
            bw=(a>thr).astype(np.uint8)*255
        n, labels, stats, cent=cv2.connectedComponentsWithStats(bw,8)
        comps=[]
        for i in range(1,n):
            x,y,w,h,area=stats[i]
            if 50<area<50000 and 10<w<1000 and 10<h<1000:
                comps.append((area,x,y,w,h))
        comps=sorted(comps, reverse=True)[:20]
        print('blur',blur,tname, comps[:10])

Python

python
from PIL import Image
import numpy as np, itertools, re, collections, math, string, zlib, bz2, lzma
img=np.array(Image.open('/mnt/data/ninja/challenge.png').convert('RGB'))
flat={ch:img[:,:,i].flatten() for i,ch in enumerate('RGB')}
pairs=[]
for a,b in itertools.permutations('RGB',2):
    data=bytes((((flat[a]&0x0F)<<4)|(flat[b]&0x0F)).tolist())
    pairs.append((f'{a}{b}',data))
    data2=bytes((((flat[a]>>4)<<4)|(flat[b]>>4)).tolist()) # high nibbles combine
    pairs.append((f'{a}{b}_hi',data2))
# across pixels maybe sequential nibbles in order
seqnibs={}
for order in itertools.permutations('RGB',3):
    nibs=np.stack([flat

tuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuvvuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuueeeeeeuuuuuuuueeeeeeeeuuuuvvvvuvvvvvvvvvvvvvvvvuuuuuuuuuuuuuuuuuuuvvvvuuuuuuuuvueeeeeeeeeeuuvvvvvvvvvvvvvvvvvvvvveeeeeeeeeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUeeeeeeeeeeeeeeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeuuuuuuuuuuuuuuuuuuuuuuvvvuuuuuuuuuuuuuuuuuuuueeeeeeeeeuvvvvvveeeeeeeeeeefeeeUUUUeeeeeffvvvvffffffffffvvvvvvvvvvvvvvffffffvvfffffvvvvvvfffffffffffffvvvvvvvffffffffvvvvvvvvvffffffeeeeeeeeeeeeeuuuvvvvvvvvvvvvvvvvvuuuvvvv')]),
  ('nib_RGB',
   'long',
   [(18862, 'G4e5c5#RV3R5#R5CT5EDb4'),
    (19822, 'XEe5cV6Se7uwWuVUeeVUsF#0'),
    (20782, 'VUV5ce6Se7uwWuVUeeVUsF#0')]),
  ('nib_RGB_rev',
   'long',
   [(789, 'dR8ws7fb&fr7ws7fb&fb&fb&fb&hr'),
    (1888, '#0$C25TC5TCFeTguVguVH'),
    (4036, '30$C2$C25TC5TC5TC$SCER$')]),
  ('hib_RGB',
   'long',
   [(0, "2s'2s'2s'2s'Ct7Ct7Ce&Se5CT5CS43S5DeGe"),
    (146, '7Sd53C43C43C43C43C43C43C43C43C43C43C43C43C43C43C43C43S5DfG'),
    (443,
     'WDSE4C44CD4CD4SE4SE4SE4SE4SE3C43C43C44SE4SE4SE4SE3B4"2#"24#C43C54TESfHt')]),
  ('hib_RGB_rev',
   'long',
   [(0,
     'WwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveVfeVfeVfeWveWveWveWveVfeVfeVfeVveWveWvegvfgvfgvfWveWveWveWveWfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVeUVeUWuUWuUVeUVeUVeUVeUVeUVeUVeUVeUVeUWufgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfWfUVeUVeUVeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUEUTEUTEUTUUUUUUUUUVeUVeUVeUVeUVeUVeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUEUTEUTUUUUUUUeUVeUVeUVeUVfeVfeVeUVeUVeUVeUVfeVfeVfeVfeVfeVfeWveWveWveWveWveWveWveWveWveWveWveWveWveWfeWveWveWveWveWveWveWfeVfeVfeVfeVfeVfeVfeVfeVfeVveVfeVfeVfeVfeVfeVfeVfeVffVfeVUUUUUUUUVeUVfeVffffffffVfffffffffffffffffffffvfgwvgwvgwvgwvgwvgwvgwvgwvgfffffVfeVfeVfeVfeVfeVfefffgffffffffgwvgwvgwvgwvgvfgwvggfffeVfeVfeVfeffeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVveWveWveWwuWwuWwuWwugwvWwuWwuWwvgwvh'),
    (827,
     'uXwuWwuWwuWwuWwuWwuWwuWwuGwtGwtGwtGwtGwtGwtGwuWwuWwuWwuWwuWwuWwtGwtGwtGwtGwtGwtGwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWveWveWwuWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveVfeVfeVfeWveWveWveWveVfeVfeVfeVveWveWvegvfgvfgvfWvegvfgvfgvfWveWveVfeVfeVfeVfeVfeWveWveWveWveWuUWuUWuUWuUWuUWuUVeUVuUWeUVeUVeUVeUVeUVuUgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgfeVfUVeUVeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUEUTEUTUUUUUUUeUVeUVeUVeUVeUVeUVeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUVeUVeUVeUVeUVfeVfeVeUVeUVeUVeUVfeVfeVfeVfeVfeVfeVveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWfeVfeVfeVfeVfeVfeVfeVvegwvgwvgwfVfeVfeVfeVfeVfeVffVfeVUUUUUUUUVeeVfeVfffffffffffffffffgvfgvffffgvfgvvgwvgwvgwvgwvgwvgwvgwvgwvggfVfeffffffffvggvfffVffffffffffffffgwvgwvgwvgffffffvfgffgfffffffffffffffffVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeWveWveWveWwuWwugwvgwvgwvgwvgwuWwuWwvgwvh'),
    (1787,
     'uWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwtGwtGwtGwtGwtWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwugwvWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveVfeVfeVfeWveWveWveWveVfeVfeVfeVfeWveWvegvfgvfWvfgvfgvfgvfgvfgvfgvfgvfgveWveWveWveWveWveWveWveWveWvegvfgvfWveWveWuUWueguUVeUVeUVeUVeUVeUWuegvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgfeVfUVeUVeUVeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUVeUVfeVfeVfeVfeVeUVeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUVeUVeUVeUVeUVeUVeUVeUVeUVeeVfeVfeVfeVfeVfeVfeVfeVfeVveWveWveWveWveWveWveWveWveWveWveWvfgvfWveWveWveWveWveWveWveWveWveWveVfeVfeVfeVfeVvegwvgwvgwfVfeVfeVfeVfeVfeVffVfeVUUUUUUeUVfeVfefffgvfgvffffffffffffffffgvvgwvgwvgwvgvfgvfgvffffffffffgvvfffffffvvgwvgwvggvffffffffffffffffffgwvgwvgwvgffffffffffffvfgvfgvfgvfgvffffffffffVfeVfeVfeVfeVfeVfeVveWvegvvgwvgwvgwvgwvgwvgwvgwvgwuWwugwvgwvh')]),
  ('nib_RBG',
   'long',
   [(18862, "'CcU62S%Sc%2S%4SE4UDc$! "),
    (19822, '8TcU6Sf5cWWuwWUeVeVUt621'),
    (20782, 'FUSe6cV5cWWuwWUeVeVUt621')]),
  ('nib_RBG_rev',
   'long',
   [(791, '5ws7fb&fr7ws7fb&fb&fb&fb&'),
    (4037, '3B#4B#4S4ES4ES4EB3ETB%1'),
    (4997, '3B#4B#4S4ES4EtEWS4ETB%1')]),
  ('hib_RBG',
   'long',
   [(0, '7#r7#r7#r7#rG4sG4sF5bV5SE4SE3C53SFEtiV'),
    (145, '7sV4S43C43C43C43C43C43C43C43C43C43C43C43C43C43C43C43C53SFFt'),
    (443,
     'uECT4CC4CD4CD5CT5CT5CT5CT5CT43C43C43C5CT5CT5CT5CT42C#"2#"C$3C43S5DTV6')]),
  ('hib_RBG_rev',
   'long',
   [(0,
     'uWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVgeVfeVfeVfuVguVguVguVgeVfeVfeVfeVguVguVgvfgvfgvfguVguVguVguVguVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeUVeUVuUWuUWeUVeUVeUVeUVeUVeUVeUVeUVeUVuegvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfguVVeUVeUVeUVUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUTEUTEUTEUUUUUUUUUUeUVeUVeUVeUVeUVeUVUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUTEUTEUUUUUUUUUVeUVeUVeUVeVfeVfeUVeUVeUVeUVeVfeVfeVfeVfeVfeVfuVguVguVguVguVguVguVguVguVguVguVguVguVguVfuVguVguVguVguVguVguVfeVfeVfeVfeVfeVfeVfeVfeVfeVgeVfeVfeVfeVfeVfeVfeVfeffeVfeUUUUUUUUeUVeVfeffffffffeffffffffffffffffffffffgvgwvgwvgwvgwvgwvgwvgwvgwvfffffeVfeVfeVfeVfeVfeVffffvffffffffvgwvgwvgwvgwvfgvgwvgffVfeVfeVfeVffVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVguVguVguWwuWwuWwuWwvgwuWwuWwugwvgw'),
    (829,
     'WwuWwuWwuWwuWwuWwuWwuWwtGwtGwtGwtGwtGwtGwtWwuWwuWwuWwuWwuWwuGwtGwtGwtGwtGwtGwtWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuVguVguWwuVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVgeVfeVfeVfuVguVguVguVgeVfeVfeVfeVguVguVgvfgvfgvfguVgvfgvfgvfguVguVgeVfeVfeVfeVfeVfuVguVguVguVguUWuUWuUWuUWuUWuUWeUVeUWuUVeUVeUVeUVeUVeUWvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgvVfeVVeUVeUVUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUTEUTEUUUUUUUUUVeUVeUVeUVeUVeUVeUVUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUeUVeUVeUVeUVeVfeVfeUVeUVeUVeUVeVfeVfeVfeVfeVfeVfeVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVfeVfeVfeVfeVfeVfeVfeVgvgwvgwvggeVfeVfeVfeVfeVfeffeVfeUUUUUUUUeUfeVfefffffffffffffffffvfgvfgfffvfgvfwvgwvgwvgwvgwvgwvgwvgwvgwvgfeVfffffffffvvgvfffeffffffffffffffvgwvgwvgwvfffffffgvffvfffffffffffffffffeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfuVguVguVguWwuWwvgwvgwvgwvgwvWwuWwugwvgw'),
    (1786,
     'WxuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuGwtGwtGwtGwtGwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwvgwuVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVgeVfeVfeVfuVguVguVguVgeVfeVfeVfeVfuVguVgvfgvfgufgvfgvfgvfgvfgvfgvfgvfgvVguVguVguVguVguVguVguVguVguVgvfgvfguVguVguUWuUgvUWeUVeUVeUVeUVeUVuUgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgvVfeVVeUVeUVeUVUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUeUVeVfeVfeVfeVfeUVeUVUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUeUVeUVeUVeUVeUVeUVeUVeUVeUfeVfeVfeVfeVfeVfeVfeVfeVfeVguVguVguVguVguVguVguVguVguVguVgufgvfguVguVguVguVguVguVguVguVguVguVgeVfeVfeVfeVfeVgvgwvgwvggeVfeVfeVfeVfeVfeffeVfeUUUUUUUVeVfeVffffvfgvfgfffffffffffffffvfwvgwvgwvgwvfgvfgvfgfffffffffvfwffffffffwvgwvgwvgvffffffffffffffffffvgwvgwvgwvfffffffffffffgvfgvfgvfgvfgfffffffffeVfeVfeVfeVfeVfeVfeVguVgvfwvgwvgwvgwvgwvgwvgwvgwvWwuWwvgwvgw')]),
  ('nib_GRB',
   'long',
   [(18861, '(CtV6SS%2e52S%2SE4SDT&1B'),
    (19824, 'V6Se5cV7uwWueVUVUe7Bc'),
    (20781, 'HUee6SV5cV7uwWueVUVUe7Bc')]),
  ('nib_GRB_rev',
   'long',
   [(792, '7ws&fb&vs7ws&fb&fb&fb&fb(vA'),
    (4038, '4B#4B#ES4ES4ES44R4%TB'),
    (4998, '4B#4B#ES4ES4WtEES4%TB')]),
  ('hib_GRB',
   'long',
   [(0, '#r7#r7#r7#r74sG4sG4bV5cU4SE4S43S5DdWV'),
    (146, 'w5cE3C43C43C43C43C43C43C43C43C43C43C43C43C43C43C43C43S5DdgY'),
    (443,
     'gDT5CC4CD4CD4CT5CT5CT5CT5CT53C43C43C4CT5CT5CT5CT53C$"2#"3$2C43C5CTE5dhG')]),
  ('hib_GRB_rev',
   'long',
   [(0,
     'wuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVfeVfeVfeVguVguVguVguVfeVfeVfeVfuVguVguVgvfgvfgvfguVguVguVguVgeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVVeUVeUWuUWuUVeUVeUVeUVeUVeUVeUVeUVeUVeUWufgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgeUVeUVeUVeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUTEUTEUTEUUUUUUUUUVeUVeUVeUVeUVeUVeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUTEUTEUUUUUUUeUVeUVeUVeUfeVfeVVeUVeUVeUVeUfeVfeVfeVfeVfeVfeVguVguVguVguVguVguVguVguVguVguVguVguVguVgeVguVguVguVguVguVguVgeVfeVfeVfeVfeVfeVfeVfeVfeVfuVfeVfeVfeVfeVfeVfeVfeVfeffeVVUUUUUUUUVeUfeVfeffffffffeffffffffffffffffffffvfwvgwvgwvgwvgwvgwvgwvgwvggffffffeVfeVfeVfeVfeVfeVfffgffffffffwvgwvgwvgwvggvfwvgwffffVfeVfeVfeVffVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfuVguVguVwuWwuWwuWwuWwvgwuWwuWwugwvgx'),
    (827,
     'WxuWwuWwuWwuWwuWwuWwuWwuWwtGwtGwtGwtGwtGwtGwtWwuWwuWwuWwuWwuWwuGwtGwtGwtGwtGwtGwtWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWguVguVwuWguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVfeVfeVfeVguVguVguVguVfeVfeVfeVfuVguVguVgvfgvfgvfguVgvfgvfgvfguVguVfeVfeVfeVfeVfeVguVguVguVguVWuUWuUWuUWuUWuUWuUVeUVuUWeUVeUVeUVeUVeUVuUgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgfVfeUVeUVeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUTEUTEUUUUUUUeUVeUVeUVeUVeUVeUVeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUVeUVeUVeUVeUfeVfeVVeUVeUVeUVeUfeVfeVfeVfeVfeVfeVfuVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVgeVfeVfeVfeVfeVfeVfeVfuVwvgwvgwvffeVfeVfeVfeVfeVfeffeVVUUUUUUUUVeVfeVfeffffffffffffffffgvfgvffffgvfgvgwvgwvgwvgwvgwvgwvgwvgwvgwfffeVffffffffgwfgffffefffffffffffffwvgwvgwvggffffffvfgffgffffffffffffffffffeVfeVfeVfeVfeVfeVfeVfeVfeVfeVguVguVguVwuWwuWwvgwvgwvgwvgwvWwuWwugwvgx'),
    (1787,
     'WwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuGwtGwtGwtGwtGwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwvgguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVfeVfeVfeVguVguVguVguVfeVfeVfeVfeVguVguVgvfgvfgufgvfgvfgvfgvfgvfgvfgvfgvVguVguVguVguVguVguVguVguVguVgvfgvfguVguVWuUWuVWvUVeUVeUVeUVeUVeUWuVgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgfVfeUVeUVeUVeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUVeUfeVfeVfeVfeVVeUVeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUVeUVeUVeUVeUVeUVeUVeUVeUVeVfeVfeVfeVfeVfeVfeVfeVfeVfuVguVguVguVguVguVguVguVguVguVguVgufgvfguVguVguVguVguVguVguVguVguVguVfeVfeVfeVfeVfuVwvgwvgwvffeVfeVfeVfeVfeVfeffeVVUUUUUUeUfeVfeVfffgvfgvffffffffffffffffgvgwvgwvgwvggvfgvfgvffffffffffgvgfffffffvgwvgwvgwfgffffffffffffffffffwvgwvgwvggffffffffffffvfgvfgvfgvfgvfffffffffffeVfeVfeVfeVfeVfeVfuVguVgvgwvgwvgwvgwvgwvgwvgwvgwvWwuWwvgwvgx')]),
  ('nib_GBR',
   'long',
   [(18863, 'GSf5R5#cU#R5#T5CT4E#a$'),
    (19823, 'XSf5cU6SgWuwWeVUUeV4r6'),
    (20783, 'VcV5Se6SgWuwWeVUUeV4r6')]),
  ('nib_GBR_rev',
   'long',
   [(792, 's7wb&fb6ws7wb&fb&fb&fb&f'),
    (4038, 'C2$C2$TC5TC5TC5CB5R$E0'),
    (4998, 'C2$C2$TC5TC5uTGTC5R$E0')]),
  ('hib_GBR',
   'long',
   [(0, "'2s'2s'2s'2s7Ct7Ct6Be6SU5CT5CC53SFDuYe"),
    (146, 'w6ST43C43C43C43C43C43C43C43C43C43C43C43C43C43C43C43C53SFDvY'),
    (443,
     'vEDSD3CD4CD4CE4SE4SE4SE4SE4S43C43C43CE4SE4SE4SE4S43B#"2##B4#C43SE4T6T')]),
  ('hib_GBR_rev',
   'long',
   [(0,
     'wuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWfeVfeVfeVveWveWveWveWfeVfeVfeVfeWveWveWvfgvfgvfgveWveWveWveWveVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVeUVeUVuUWuUWeUVeUVeUVeUVeUVeUVeUVeUVeUVuegvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgvUVeUVeUVeUVUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUTEUTEUTEUUUUUUUUUeUVeUVeUVeUVeUVeUVUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUTEUTEUUUUUUUUVeUVeUVeUVfeVfeVeUVeUVeUVeUVfeVfeVfeVfeVfeVfeVveWveWveWveWveWveWveWveWveWveWveWveWveWveVveWveWveWveWveWveWveVfeVfeVfeVfeVfeVfeVfeVfeVfeWfeVfeVfeVfeVfeVfeVfeVfeffeVeUUUUUUUUeUVfeVfeffffffffefffffffffffffffffffffgwvgwvgwvgwvgwvgwvgwvgwvgvffffffeVfeVfeVfeVfeVfeVfffvffffffffwvgwvgwvgwvgvfgwvgwffffVfeVfeVfeVffVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeWveWveWwuWwuWwuWwuWwvgwuWwuWwugwvg'),
    (829,
     'uWwuWwuWwuWwuWwuWwuWwuWwtGwtGwtGwtGwtGwtGwtWwuWwuWwuWwuWwuWwuGwtGwtGwtGwtGwtGwtWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWveWveWwuWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWfeVfeVfeVveWveWveWveWfeVfeVfeVfeWveWveWvfgvfgvfgveWvfgvfgvfgveWveWfeVfeVfeVfeVfeVveWveWveWveWuUWuUWuUWuUWuUWuUWeUVeUWuUVeUVeUVeUVeUVeUWvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfVfUVeUVeUVUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUTEUTEUUUUUUUUVeUVeUVeUVeUVeUVeUVUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUeUVeUVeUVeUVfeVfeVeUVeUVeUVeUVfeVfeVfeVfeVfeVfeVfeWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveVfeVfeVfeVfeVfeVfeVfeWwvgwvgwfgfeVfeVfeVfeVfeVfeffeVeUUUUUUUUeeVfeVfeffffffffffffffffvfgvfgfffvfgvvgwvgwvgwvgwvgwvgwvgwvgwvgwfffeVfffffffvfwvfffffefffffffffffffwvgwvgwvgvfffffffgvffvffffffffffffffffffeVfeVfeVfeVfeVfeVfeVfeVfeVfeVveWveWveWwuWwuWwvgwvgwvgwvgwvWwuWwugwvg'),
    (1786,
     'uXwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuGwtGwtGwtGwtGwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwvgveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWfeVfeVfeVveWveWveWveWfeVfeVfeVfeVveWveWvfgvfgvegvfgvfgvfgvfgvfgvfgvfgvfWveWveWveWveWveWveWveWveWveWvfgvfgveWveWuUWueWuVWeUVeUVeUVeUVeUVueWvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfVfUVeUVeUVeUVUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUeUVfeVfeVfeVfeVeUVeUVUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUeUVeUVeUVeUVeUVeUVeUVeUVeeVfeVfeVfeVfeVfeVfeVfeVfeVfeWveWveWveWveWveWveWveWveWveWveWvegvfgveWveWveWveWveWveWveWveWveWveWfeVfeVfeVfeVfeWwvgwvgwfgfeVfeVfeVfeVfeVfeffeVeUUUUUUUVfeVfeVfffvfgvfgfffffffffffffffvvgwvgwvgwvgvfgvfgvfgfffffffffvvgfffffffvgwvgwvgwvfffffffffffffffffffwvgwvgwvgvfffffffffffffgvfgvfgvfgvfgffffffffffeVfeVfeVfeVfeVfeVfeWveWvvgwvgwvgwvgwvgwvgwvgwvgwvWwuWwvgwvg')]),
  ('nib_BRG',
   'long',
   [(11776, "Q%3c63c'2s'2s'Ru'2s'"),
    (12736, "Q%3c63c62s'2s'Ru'2s'"),
    (18861, 'H$s6SV#R55b5#R5CT5CTT6"A')]),
  ('nib_BRG_rev',
   'long',
   [(791, 'S7ws&fb&vs7ws&fb&fb&fb&fb&Hq'),
    (1889, '#2$CC5TC5TTFeVguVguTH'),
    (4037, '32$C2$CC5TC5TC5T24T$ER')]),
  ('hib_BRG',
   'long',
   [(0, "s'2s'2s'2s'2t7Ct7Cd6Re5ST5CT43S53dGT"),
    (145, '7se5CC43C43C43C43C43C43C43C43C43C43C43C43C43C43C43C43S53dGd'),
    (443,
     'eTE4CD3CD4CD4SE4SE4SE4SE4SE4C43C43C43SE4SE4SE4SE4C4#2#"2$#B43C53SEDe8d')]),
  ('hib_BRG_rev',
   'long',
   [(0,
     'uWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWweWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveVfeVfeVfeWveWveWveWveVfeVfeVfeVveWveWvfgvfgvfgveWveWveWveWveWfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfUVeUVeUWuUWuUVeUVeUVeUVeUVeUVeUVeUVeUVeUgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgveWeUVeUVeUVeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUTEUTEUTEUUUUUUUUUUUVeUVeUVeUVeUVeUVeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUTEUTEUUUUUUUUUeUVeUVeUVeeVfeVfUVeUVeUVeUVeeVfeVfeVfeVfeVfeVfeWveWveWveWveWveWveWveWveWveWveWveWveWveWfeWveWveWveWveWveWveWfeVfeVfeVfeVfeVfeVfeVfeVfeVveVfeVfeVfeVfeVfeVfeVfeffeVfUVUUUUUUUUVeeVfeffffffffeffffffffffffffffffffffvvgwvgwvgwvgwvgwvgwvgwvgwfgffffeVfeVfeVfeVfeVfeVfffffgfffffffvgwvgwvgwvgwfgvvgwvgffVfeVfeVfeVffVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVveWveWvuWwuWwuWwuWwvgwuWwuWwugwvgwvh'),
    (828,
     'uXwuWwuWwuWwuWwuWwuWwuWwtGwtGwtGwtGwtGwtGwtWwuWwuWwuWwuWwuWwuGwtGwtGwtGwtGwtGwtWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWweWveWvuWweWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveVfeVfeVfeWveWveWveWveVfeVfeVfeVveWveWvfgvfgvfgveWvfgvfgvfgveWveWveVfeVfeVfeVfeVfeWveWveWveWvUWuUWuUWuUWuUWuUWuUVeUVuUWeUVeUVeUVeUVeUVufgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfWfeVeUVeUVeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUTEUTEUUUUUUUUUeUVeUVeUVeUVeUVeUVeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUVeUVeUVeUVeeVfeVfUVeUVeUVeUVeeVfeVfeVfeVfeVfeVfeVveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWfeVfeVfeVfeVfeVfeVfeVvvgwvgwvgveVfeVfeVfeVfeVfeffeVfUVUUUUUUUUVfeVfeffffffffffffffffffgvfgvffffgvfgwvgwvgwvgwvgwvgwvgwvgwvgwvgfeVfffffffffgvggfffeffffffffffffffvgwvgwvgwfgffffffvfgffgffffffffffffffffeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeWveWveWvuWwuWwvgwvgwvgwvgwvWwuWwugwvgwvh'),
    (1788,
     'uWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuGwtGwtGwtGwtGwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwvgweWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveWveVfeVfeVfeWveWveWveWveVfeVfeVfeVfeWveWvfgvfgvegvfgvfgvfgvfgvfgvfgvfgvfWveWveWveWveWveWveWveWveWveWvfgvfgveWveWvUWuUWvVWuUVeUVeUVeUVeUVeUWvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfWfeVeUVeUVeUVeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUVeeVfeVfeVfeVfUVeUVeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUVeUVeUVeUVeUVeUVeUVeUVeUVfeVfeVfeVfeVfeVfeVfeVfeVfeVveWveWveWveWveWveWveWveWveWveWvegvfgveWveWveWveWveWveWveWveWveWveWveVfeVfeVfeVfeVvvgwvgwvgveVfeVfeVfeVfeVfeffeVfUVUUUUUUeeVfeVfffffgvfgvffffffffffffffffgwvgwvgwvgwfgvfgvfgvffffffffffgwffffffffwvgwvgwvggffffffffffffffffffvgwvgwvgwfgffffffffffffvfgvfgvfgvfgvfffffffffeVfeVfeVfeVfeVfeVfeVveWvfgwvgwvgwvgwvgwvgwvgwvgwvWwuWwvgwvgwvh')]),
  ('nib_BGR',
   'long',
   [(11776, '!R63c63r7#r7#rW%r7#r'),
    (12736, '!R63c63c7#r7#rW%r7#r'),
    (18863, '75ce%2S6RS%2SE4SE4E2b')]),
  ('nib_BGR_rev',
   'long',
   [(788, '$b4Xs7wb&fb6ws7wb&fb&fb&fb&fb'),
    (1888, '2 #4B4ES4ESEVdeWveWvEX'),
    (4036, '30#4B#4B4ES4ES4ES#DSB%T')]),
  ('hib_BGR',
   'long',
   [(0, 'r7#r7#r7#r7#sG4sG4cF%cU5SE4SD3S53dGE'),
    (146, '7cU4C43C43C43C43C43C43C43C43C43C43C43C43C43C43C43C43S53dGF'),
    (443,
     'VTECD43D4CD4CT5CT5CT5CT5CT5CC43C43C43T5CT5CT5CT5CC422#"2$2C$3C53T5DcXF')]),
  ('hib_BGR_rev',
   'long',
   [(0,
     'WwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVfeVfeVfeVguVguVguVguVfeVfeVfeVfuVguVgufgvfgvfgvVguVguVguVguVgeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeUVeUVeUWuUWuUVeUVeUVeUVeUVeUVeUVeUVeUVeUgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgvVWeUVeUVeUVeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUEUTEUTEUTUUUUUUUUUUVeUVeUVeUVeUVeUVeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUEUTEUTUUUUUUUUeUVeUVeUVeVfeVfeUVeUVeUVeUVeVfeVfeVfeVfeVfeVfeVguVguVguVguVguVguVguVguVguVguVguVguVguVgeVguVguVguVguVguVguVgeVfeVfeVfeVfeVfeVfeVfeVfeVfuVfeVfeVfeVfeVfeVfeVfeVffVfeUVUUUUUUUUVeVfeVffffffffVffffffffffffffffffffffvgwvgwvgwvgwvgwvgwvgwvgwvfgffffVfeVfeVfeVfeVfeVfeffffgfffffffgwvgwvgwvgwvfgvgwvggfffeVfeVfeVfeffeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVfuVguVguWwuWwuWwuWwugwvWwuWwuWwvgwvgx'),
    (828,
     'WxuWwuWwuWwuWwuWwuWwuWwuGwtGwtGwtGwtGwtGwtGwuWwuWwuWwuWwuWwuWwtGwtGwtGwtGwtGwtGwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuVguVguWwuVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVfeVfeVfeVguVguVguVguVfeVfeVfeVfuVguVgufgvfgvfgvVgufgvfgvfgvVguVguVfeVfeVfeVfeVfeVguVguVguVguUWuUWuUWuUWuUWuUWuUVeUVuUWeUVeUVeUVeUVeUVufgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgeVVeUVeUVeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUEUTEUTUUUUUUUUeUVeUVeUVeUVeUVeUVeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUVeUVeUVeUVeVfeVfeUVeUVeUVeUVeVfeVfeVfeVfeVfeVfeVfuVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVgeVfeVfeVfeVfeVfeVfeVfugwvgwvggvVfeVfeVfeVfeVfeVffVfeUVUUUUUUUUfeVfeVffffffffffffffffffgvfgvffffgvfwvgwvgwvgwvgwvgwvgwvgwvgwvggfVfefffffffvfgwffffVffffffffffffffgwvgwvgwvfgffffffvfgffgffffffffffffffffVfeVfeVfeVfeVfeVfeVfeVfeVfeVfeVguVguVguWwuWwugwvgwvgwvgwvgwuWwuWwvgwvgx'),
    (1788,
     'WwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwtGwtGwtGwtGwtWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwuWwugwvVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVguVfeVfeVfeVguVguVguVguVfeVfeVfeVfeVguVgufgvfgvVgvfgvfgvfgvfgvfgvfgvfgvfguVguVguVguVguVguVguVguVguVgufgvfgvVguVguUWuUgueWuUVeUVeUVeUVeUVeUgufgvfgvfgvfgvfgvfgvfgvfgvfgvfgvfgeVVeUVeUVeUVeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUVeVfeVfeVfeVfeUVeUVeUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUVeUVeUVeUVeUVeUVeUVeUVeUfeVfeVfeVfeVfeVfeVfeVfeVfeVfuVguVguVguVguVguVguVguVguVguVguVgvfgvVguVguVguVguVguVguVguVguVguVguVfeVfeVfeVfeVfugwvgwvggvVfeVfeVfeVfeVfeVffVfeUVUUUUUUeVfeVfeffffgvfgvffffffffffffffffwvgwvgwvgwvfgvfgvfgvffffffffffwvfffffffvvgwvgwvgwfffffffffffffffffffgwvgwvgwvfgffffffffffffvfgvfgvfgvfgvfffffffffVfeVfeVfeVfeVfeVfeVfuVgufwvgwvgwvgwvgwvgwvgwvgwvgwuWwugwvgwvgx')])],
 48)

Python

python
import itertools, re, numpy as np
from PIL import Image
img=np.array(Image.open('/mnt/data/ninja/challenge.png').convert('RGB'))
flat={ch:img[:,:,i].flatten() for i,ch in enumerate('RGB')}

streams={}
# byte streams from nibble combinations
for a,b in itertools.permutations('RGB',2):
    streams[f'{a}{b}_low']=bytes((((flat[a]&0x0F)<<4)|(flat[b]&0x0F)).tolist())
for order in itertools.permutations('RGB',3):
    nibs=np.stack([flat[ch]&0x0F for ch in order],axis=1).reshape(-1)
    data=bytes(((nibs[0::2]<<4)|nibs[1::2]).tolist())
    streams[f'nibs_{order}']=data
# bit streams from low bitplanes
for order in itertools.permutations('RGB',3):
    seq=np.stack([flat[ch] for ch in order],axis=1).reshape(-1)
    for bit in range(4): # low bits only
        bits=((seq>>bit)&1).astype(np.uint8)
        for off in range(8):
            n=len(bits[off:])//8*8
            data=np.packbits(bits[off:off+n]).tobytes()
            streams[f'bits_{order}_b{bit}_o{off}']=data

flag_re = re.compile(rb'UMASS\{[^}]{0,100}\}')
found=[]
for name,data in list(streams.items()):
    candidates=[('orig',data),('rev',data[::-1])]
    for mode,d in candidates:
        m=flag_re.search(d)
        if m: found.append((name,mode,'plain',m.group()))
        for key in range(256):
            x=bytes([b^key for b in d[:200000]])  # enough
            if b'UMASS{' in x:
                idx=x.find(b'UMASS{')
                found.append((name,mode,'xor',key,idx,x[idx:idx+80]))
                break
len(found), found[:20]

思考

undefined

思考

undefined

Python

python
from PIL import Image
import numpy as np, itertools, math
img=np.array(Image.open('/mnt/data/ninja/challenge.png').convert('RGB'))
low=[(img[:,:,i]&0x0F).astype(float) for i in range(3)]
# cross-correlation shifts within +-5
for (i,a),(j,b) in itertools.combinations(list(enumerate(low)),2):
    best=None
    for dy in range(-3,4):
        for dx in range(-3,4):
            aa=a[max(0,dy):a.shape[0]+min(0,dy), max(0,dx):a.shape[1]+min(0,dx)]
            bb=b[max(0,-dy):b.shape[0]+min(0,-dy), max(0,-dx):b.shape[1]+min(0,-dx)]
            corr=np.corrcoef(aa.flatten(),bb.flatten())[0,1]
            if best is None or corr>best[0]:
                best=(corr,dx,dy)
    print(i,j,best)

思考

undefined

思考

Connecting the quote scene to hidden image

The transcript at 04:21 says, "They're surrounding you, perfectly hidden, ready to strike," and that might be key. Since the original image is from a later, crystal-themed fight scene, maybe the hidden image is from this earlier jungle scene. The background could be jungle shapes, and perhaps the ninjas form part of the hidden scene. I’ll focus on the flag phrase from the quote: “surrounding_you_perfectly_hidden_ready_to_strike.”

Choosing the flag and considering possibilities

There are two variants for the phrase: “they are” vs “they’re,” and I think the flag should use the one without the apostrophe. So the flag might be UMASS{surrounding_you_perfectly_hidden_ready_to_strike} or UMASS{theyre_surrounding_you_perfectly_hidden_ready_to_strike}. However, the hidden message could be something a ninja says, like "ca-kaw! ca-kaw!" based on the CTF style. It's likely UMASS{perfectly_hidden_ready_to_strike}, as it's concise. I'll also consider other possible flag formats based on challenge context.

Thought for 19m 43s

回答

Mình tách ảnh theo nibble thấp và thử các phép biến đổi đơn giản. Ảnh ẩn lộ ra đúng “little ninja-nerds” đang núp quanh cảnh, nên clue trỏ tới câu đáp trong phim:

UMASS{theyre_surrounding_you_perfectly_hidden_ready_to_strike}

Nếu server không nhận, biến thể mình sẽ thử ngay sau đó là:

UMASS{they_are_surrounding_you_perfectly_hidden_ready_to_strike}

问题

问题

Python

Python

Python

思考

Python

Python

思考

思考

Python

Python

Python

Python

思考

思考

Python

思考

思考

回答

分享这个问答