ImageProcessing-multithreading-python

Grayscale conversion of multiple images using openCV. Completing the task faster with multithreading in python.

Key Concept

Multithreading - Multithreading is defined as the ability of a processor to execute multiple threads concurrently.

Task Performed

• For this experiment blocks of 100, 200, 300, 400 and 500 image files were taken as input.
• rgb_to_grayscale() function defined in gray_scale.py and multithreading_gray_scale.py was used to convert image in bgr format to grayscale using cvtColor() function of openCV module.
• In gray_scale.py program task was formed sequentially but in multithreading_gray_scale.py the same task was performed with multithreading. To create threads ThreadPoolExecutor (from concurrent.futures module) was used.
• In main.py program task is performed multiple times varying number of threads each time. This is done to find the optimal number of threads which will perform the task in less time.
• plot_thread.py file is used to plot number of thread vs time graph which is result of main.py program (Report/report).
• plot_graph.py file is used to plot number of files vs time double line graph comparing the result of task performed with multithreading and without multithreading.

Observations

Table 1: Comparing execution time of task with varying number of threads

Number of Active Threads	Time Taken (s)
1	806.7504532
2	544.4976104000001
3	548.5975074
4	545.6509420000002
5	552.1994285999999
6	360.4934754000001
7	366.9670584999999
8	364.5329006000002
9	364.0383883000004
10	363.9121553999994
11	366.98776720000023
12	364.5366519999998
13	364.0464326000001
14	362.63782359999914
15	364.70052849999956
16	364.47398650000014
17	362.6452294999999
18	361.66494629999943
19	361.17374199999995
20	361.59370090000084

Graph 1:

Optimal Number of threads = 6

Table 2: Comparing execution time of capitalizing file text with multithreading and without multithreading while increasing the number of files by 100

Number of threads used = 13

Number of Files	Time Taken Without Multithreading (s)	Time Taken With Multithreading (s)
100	32.69129072803762	15.775374412536621
200	96.38587993835710	35.1330668926239
300	161.58046132345781	68.54701542854309
400	260.4525678123675	108.12210512161255
500	259.7357916341918	136.66275644302368

Graph 2:

Findings:

From graph 2 it is evident that the task performed using multithtreading takes less time.
Percent decrease in time : 55.34 %