Today?s world is evolving very fast. In this modern era, Gender Classification is getting huge importance because gender information can be used by expert and intelligent systems that are part of healthcare, smart spaces, and biometric-based access control applications. For example, operations of in
Gender Classification in human using GAIT features
Today’s world is evolving very fast. In this modern era, Gender Classification is getting huge importance because gender information can be used by expert and intelligent systems that are part of healthcare, smart spaces, and biometric-based access control applications. For example, operations of intelligent systems in a smart space can be customized based on gender information to provide an enhanced user experience. Classification of gender can be done in various ways, but we have chosen one of the most convenient methods of classification using android application through different methods based on GAIT features. In this application, we are considering the walking speed of humans by taking the readings from Accelerometer and Gyroscope sensors. As well as we will be using Camera for Face Recognition by comparing the pictures with the datasets and, we will be collecting statistical data including head, waist, chest, hairstyle, back, legs, etc. by means of user inputs. As we have seen that the previous researches were usually based on a single method, due to this the results concluded that were not consistent in extreme cases. Since our project is a combination of various methods therefore the results obtained from this project are more accurate than other existing methods and this thing will be beneficial for companies and other intelligent systems.
We are going to develop an android application for our project. In this application, we will be collecting data using three different methods. i.e., built-in Accelerometer and Gyroscope Sensors, Face Recognition, and Statistical data acquired by users input.
So, In the first technique, we will take the readings from both sensors (Accelerometer and Gyroscope) in X, Y, and Z directions based on the walking speed of users by putting the mobile phone in the pocket. After collecting the readings, we will process it by applying Normalization and Gait Cycle Extraction and Histogram Binnig techniques. Now, In the second technique, we will be capturing multiple face images of a user from different angles and then we will apply the MobileNet/inception v3 model to compare them with the train images which will be done by applying a Convolutional Neural Networks on the big, qualified datasets. Afterward, we will also collect some parameters including hairstyle, waist length, height, thickness of legs, head, age, etc. from users, and then we will be comparing them with the already trained data set which will be made available from a survey or other resources.
After processing all the methods and techniques we will be classifying gender based on the results of all three methods by comparing the accuracy of these methods.
The technical deliverable of this project will be an android based software by which gender can be classified.
Initially, research and survey will be conducted, and work will be done to collect data.
| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| smart device | Equipment | 1 | 35000 | 35000 |
| Highly Qualified data sets | Equipment | 2 | 10000 | 20000 |
| Survey Conduct Charges | Miscellaneous | 7 | 1000 | 7000 |
| Thesis Printing and Binding Cost | Miscellaneous | 1 | 2000 | 2000 |
| Total in (Rs) | 64000 |
This project is mainly related with the applications of drones. The main purpose is to det...
Drones as every new emerging technology has introduced many threats and risks to human soc...
In order to assist and improvise the living conditions of the physically-challenged or age...
In today?s world the majority transport through cars. Talking about this we can see many p...
The goal of this project is to introduce and present a machine learning application that a...