Replacing a for loop with a time efficient alternative ?

So I am making my own algorithm to implement binary huffman encoding for grayscale images, my code works very well but the encoding process for loop takes too much time if the image has many levels, it goes like this:

(The S stands for symbols and the Levels is the input image reshaped into a vector, dict is my formed codeword dictionary which is similar to using huffmandict)

%Encoding the stream:
Encoded_Stream = [];
Decoding_Cell = {};
for i=1:length(Levels)
    for j=1 : length(S)
        if Levels(i) == dict{j,1}
            Encoded_Stream = [Encoded_Stream dict{j,2}];
            Decoding_Cell{i,1} = dict{j,2};
        end
    end
end

Is there any faster solution to find the corresponding codeword in the dictionary without using this second for loop

[Mehmed Saad]

also i think you can remove the 2nd for loop completly. you have to play with dict and cells

[Mahmoud Emad]

Yes I figured that the second loop is the actual problem but I can't come up with an alternative solution

[Mehmed Saad]

can you provide the cell array dict here

[Mahmoud Emad]

109	[0,0]
16	[0,1]
0	[1,0]
64	[1,1,0]
110	[1,1,1,0]
108	[1,1,1,1,0]
35	[1,1,1,1,1,1,0,0]
88	[1,1,1,1,1,1,1,1,1]
19	[1,1,1,1,1,0,0,0,1,0]
105	[1,1,1,1,1,0,0,0,1,1]
61	[1,1,1,1,1,0,0,1,0,1]
29	[1,1,1,1,1,0,1,1,1,0]
75	[1,1,1,1,1,0,1,1,1,1]
48	[1,1,1,1,1,1,0,1,0,1]
15	[1,1,1,1,1,1,0,1,1,0]
94	[1,1,1,1,1,1,0,1,1,1]
100	[1,1,1,1,1,1,1,0,0,0]
24	[1,1,1,1,1,1,1,0,1,1]
8	[1,1,1,1,1,1,1,1,0,0]
1	[1,1,1,1,1,1,1,1,0,1]
4	1x11 double
12	1x11 double
6	1x11 double
26	1x11 double
40	1x11 double
10	1x11 double
2	1x11 double
53	1x11 double
14	1x11 double
70	1x11 double
13	1x11 double
3	1x11 double
81	1x11 double
5	1x12 double
36	1x12 double
62	1x12 double
11	1x12 double
89	1x12 double
9	1x12 double
7	1x12 double
17	1x12 double
20	1x12 double
106	1x12 double
59	1x12 double
76	1x12 double
96	1x12 double
37	1x12 double
25	1x13 double
49	1x13 double
107	1x13 double
42	1x13 double
69	1x13 double
31	1x13 double
50	1x13 double
51	1x13 double
56	1x13 double
30	1x13 double
90	1x13 double
22	1x13 double
101	1x13 double
21	1x13 double
41	1x13 double
54	1x13 double
71	1x13 double
82	1x13 double
95	1x13 double
72	1x13 double
77	1x13 double
102	1x13 double
84	1x13 double
99	1x13 double
33	1x14 double
85	1x14 double
104	1x14 double
45	1x14 double
73	1x14 double
27	1x14 double
38	1x14 double
91	1x14 double
34	1x14 double
78	1x14 double
97	1x14 double
86	1x15 double
58	1x15 double
74	1x15 double
46	1x15 double
98	1x15 double
39	1x16 double
92	1x16 double
18	1x16 double
28	1x16 double
103	1x16 double
23	1x17 double
68	1x17 double
32	1x18 double
43	1x19 double
44	1x19 double
47	1x18 double
52	1x18 double
55	1x18 double
57	1x18 double
60	1x18 double
63	1x18 double
65	1x18 double
66	1x18 double
67	1x18 double
79	1x18 double
80	1x18 double
83	1x18 double
87	1x18 double
93	1x18 double

First column are the symbols(Levels) and the second is the codeword represented in a vector

[Mehmed Saad]

i mean in the mat format

[Mahmoud Emad]

I don't have it in a mat format actually since the length of the codeword is variable so it had to be in a cell

[Mehmed Saad]

just right click the codeword variable and save as mat file

xyz.mat

attach tha xyz.mat here

[Mahmoud Emad]

I came up with an idea finally to encode the stream without using an array and a cell that change size each loop iteration. Thank you very much for your consideration it helped me come up with the idea :D

[Mehmed Saad]

Cheers