Gender Classification using SVM With Flask

International Journal of Electrical, Electronics and Computers
Vol-6, Issue-3 | May-Jun, 2021
Available: https://aipublications.com/ijeec/
Peer-Reviewed Journal
ISSN: 2456-2319
https://dx.doi.org/10.22161/eec.63.6 43
Gender Classification using SVM With Flask
Akhil Maheshwari, Dolly Sharma, Ritik Agarwal, Shivam Singh, Loveleen Kumar
Department of CSE, Global Institute of Technology, Jaipur, India
Received: 01 Jun 2021; Accepted: 16 Jun 2021; Date of Publication: 23 Jun 2021
©2021 The Author(s). Published by Infogain Publication. This is an open access article under the CC BY license
(https://creativecommons.org/licenses/by/4.0/).
Abstract— The main objective of this work is the uniting and streamlining of an automatic face
detection application and recognition system for video indexing applications. Human identification
means the classification of gender which can increase the identification accuracy. So, accurate gender
classification algorithms may increase the accuracy of the applications and can reduce its complexity.
But, in some applications, some challenges are there such as rotation, gray scale variations that may
reduce the accuracy of the application. The main goal of building this module is to understand the
values in image, pattern, and array processing with OpenCV for effective processing faces for building
pipe-lining, SVM models.
Keywords— Gender classification, Human Recognition, Facial Images, OpenCv, Image Processing.
I. INTRODUCTION
Human's face identification is a major hallmark in
discernible machine learning and image processing
systems to identify desired target. A human face tote
discriminative and separable information consisting
gender, age, ethnicity, etc. Face information is applicable
in many cases such as human-computer interaction, image
retrieval, biometric authentication, drivers monitoring,
human-robot interaction, sport competition senses
analysis, video summarizing, and image/video indexing.
Basically we are building a image processing application
which is face recognition and it also try to detect the facial
features and classify the gender i.e male or female. The
process is starting by gathering the data. As the data is
gathered we are going to arrange the highly unstructured
data in a structured form after that we are learning image
techniques using OpenCV and get mathematical concept
behind the image. For image analysis we require
techniques for preprocessing our data, extract feature from
image like computing eigen images using PCA(Principal
Component Analysis) and Single Value Decomposition.
With the eigen images we can learn to test our Machine
Learning model, and the model before deploying it. After
Machine Learning model is ready we are going to learn to
develop web server gateway in flask by using HTML, CSS
and BOOTSTRAP in front-end part and PYTHON for
back-end part. The below image shows the steps for
building the application.
II. METHODOLOGY
In this Web App we will be working on input image of
human beings. The uploaded image must be clean and
clear to extract the human facial features from image taken
as Input for the processing. We tested the project only on
the human faces. The following methods is used in this
technology:-
1. Image Uploading
2. RGB/BGR to gray scale
3. Image Data Preprocessing and Analysis
4. Predictive Model
5. Displaying result
III. IMAGE UPLOADING
This is initial phase of the project. We will be uploading
the image to the project.
It is a normal image of human being. The image type must
be JPEG or PNG. We have used HTML form to upload the
image.

Akhil Maheshwari at al. Gender Classification using SVM With Flask
ISSN: 2456-2319
Fig. 1: Uploaded Image
IV. RGB/BGR to GREY SCALE
The image is passed to pipeline model which is integrated
with flask. Understanding Data is must within the project
life cycle. We have to understand the data starting from
domain knowledge.
We will work on the image taken as input. It can be in
RGB or BGR format. The digital images are used to
represented in multi-dimension arrays. We will be
converting that image into gray scale using cv2 library.
Gray scale image is two dimension array. The function
used to convert image into gray scale is given below:-
if color == 'bgr':
gray = cv2.cvtColor(imageFile,
cv2.COLOR_BGR2GRAY)
else:
gray = cv2.cvtColor(imageFile,
cv2.COLOR_RGB2GRAY)
Fig. 2: Image Conversion
V. ImAGE DATA ANLYSIS &PREPROCESSING
In this phase we will be working on the gray scale image
which is extracted from the input image which is obtained
from above methods. There are four kinds of analysis that
need to perform on the extracted image for better decision
making and they are listed below:
Defining Transformation: In this step we will be finding
the shape of the image. We will be finding the optimal size
from the histogram and box-plot.
Hazed Removal (Noise Remove): In this step we will be
removing noise from the image by adjusting light
intensity, estimating transmission map and clarifying
image.
Image Segmentation: In this step we will be partitioning
the image into multiple segments (sets of pixels, also
known as image objects). The goal of segmentation is to
simplify and/or change the representation of an image into
something that is more meaningful and easier to analyze.
Image segmentation is typically used to locate objects and
boundaries (lines, curves, etc.) in images. More precisely,
image segmentation is the process of assigning a label to
every pixel in an image such that pixels with the same
label share certain characteristics.
Object Detection: Face detection is a computer technology
being used in a variety of applications that identifies
human faces in digital images.
Python library OpenCV will be used for face recognition.
We will be detecting the face from the gray scale image
using cv2 library. We will be performing further more
methods on that face detected using the given function:-
faces = haarcascade.detectMultiScale(gray,1.5,3)
Fig. 3: Cropped Image
VI. FEATURE EXTRACTION
We cannot directly detect faces that were cleaned and
extracted faces using object detection techniques with
Machine Learning models. We will need to extract the
features from the face. The Feature extraction
techniques we used are an eigen images approach.
Eigen face
Fig. 4: Eigen Image

ISSN: 2456-2319
An eigen face is the name given to a set of eigen vectors
when used in the computer vision problem of human face
recognition. We will be deriving the eigen vectors from the
covariance matrix of the probability distribution over the
high-dimensional vector space of face images. This
produces dimension reduction by allowing the smaller set
of basis images to represent the original training images.
Classification can be achieved by comparing how faces are
represented by the basis set.
VII. PREDICTIVE MODEL
We have trained a machine learning model with the eigen
image data. There are many machine learning models are
there but for the image we will be using SVM.
In this phase we will be detecting gender of each face
using trained SVM model. Now there each Face is
recognized individually in this phase.
Fig. 5: Feeding Images
VIII. DISPLAYING MODEL
This is the final phase of the project. We have performed
all the method. We will be displaying Gender of the Image
and the predicted score.
Fig. 6: Producing Output
Fig. 7: Interface for Upload
Fig. 8: Data Flow
IX. CONCLUSION
In this Gender classification project we made software
which detects Gender using SVM model. It will be finding
the face of the human and predicting the gender. Though
we have tried to make efficient software but there are some
conditions for this software to work:-
• Image of the human faces should be clean and clear.
• Image quality should be high.
X. FUTURE WORK
On the future direction, results that are Good for

ISSN: 2456-2319
gender recognition as well as years opinion can
continue to be received utilizing transfer learning
strategies with expansion in reliability. Combos of
fusions as well as datasets of attributes might be what is on
the horizon for the development.
ACKNOWLEDGEMENTS
An acknowledgement section may be presented after the
conclusion, if desired.
REFERENCES
[1] [online] Available: https://www.engpaper.com/face-
recognition-2018.htm.J. Clerk Maxwell, A Treatise on
Electricity and Magnetism, 3rd ed., vol. 2. Oxford:
Clarendon, 1892, pp.68–73.
[2] Samuel Manoharan, "Image Detection Classification and
Recognition for Leak Detection in Automobiles", Journal of
Innovative Image Processing (JIIP), vol. 1, no. 02, pp. 61-
70, 2019.K. Elissa, “Title of paper if known,” unpublished.
[3] [online] Available:
https://www.semanticscholar.org/paper/Face-Detection-
using-Digital-Image-Processing-Jindal-
Gupta/a0a9390e14beb38c504473c3adc857f8faeaebd2
[4] Md. Khalid Rahmani, M.A. Ansari and Amit Goel, "An
efficient indexing algorithm for CBIR", Proceedings - 2015
IEEE International Conference on Computational
Intelligence and Communication Technology CICT 2015,
pp. 73-77, 2015.
[5] Cao Qiong, Shen Li, Xie Weidi, Omkar M Parkhi and
Andrew Zisserman, "VGGFace 2: A dataset for recognising
faces across pose and age", IEEE Conference on Automatic
Face and Gesture Recognition, 2018.
[6] Junjue Wang, Brandon Amos, Anupam Das, Padmanabhan
Pillai, Norman Sadeh and Mahadev Satyanarayanan,
"Enabling Live Video Analytics with a Scalable and
Privacy-Aware Framework", ACM Transactions on
Multimedia Computing Communications and Applications
(TOMM), vol. 14, no. 3s, pp. 64, 2018.
[7] Arsha Nagrani, Samuel Albanie and Andrew
Zisserman, Seeing Voices and Hearing Faces: Cross-modal
biometric matching, 2018, [online] Available:

Gender Classification using SVM With Flask

More Related Content

What's hot (20)

Similar to Gender Classification using SVM With Flask (20)

More from AI Publications (20)

Recently uploaded (20)

Gender Classification using SVM With Flask