Interdisciplinary - Better Life Style

Breast cancer is the one of the most common cancer among women and a major cause of increased mortality rate worldwide. The mitoses give an important indicator of the invasive breast carcinoma and hence are of great significance in diagnosis and treatment. However, the manual annotation by histologists is time-consuming and subjective with limited reproducibility. To this end, the development of automatic detection methods is essential for improving the efficiency and reliability of pathological examination. However, automatic mitosis detection in breast histology images is a very challenging task for several reasons. As the mitosis is characterized by a large variety of shape con- figurations, which are related to the high variation of biological structures. Also due to resemblance to other cell types such as apoptotic cells which carry similar morphological appearance with mitosis resulting in lots of false positives in the detection process. Further, the different conditions of histology image acquisition process, including sampling, cutting, staining, and digitalizing, increase the variabilities of mitosis appearance. This is common when the tissue samples are acquired from different patients or at different time slots. The number of mitoses per tissue area gives an important aggressiveness indication of the invasive breast carcinoma. However, automatic mitosis detection in histology images remains a challenging problem. Traditional methods either employ hand-crafted features to discriminate mitoses from other cells or construct a pixel-wise classifier to label every pixel in a sliding window way. While the former suffers from the large shape variation of mitoses and the existence of many mimics with a similar appearance, the slow speed of the later prohibits its use in clinical practice. The number of mitotic figures visible in histology sections is an important indicator for cancer screening and assessment. Normally, the count is performed manually by histologists, but automating the process could reduce its time and costs (thus making it more accessible), minimize errors, and improve the comparability of results obtained in different labs. Mitosis detection is very hard. In fact, mitosis is a complex process during which a cell nucleus undergoes various transformations.

INTERNSHIPS ARE CLOSED FOR THIS PROJECT.

Principal Investigator:

Ms. Chetna Kaushal Phd Scholar, CSE

 Chitkara University

Funding- Internal        Project Start Date -

A number of resources are available on the internet, people are habitual of using them without proper citation of the original source or without giving credit to the author.The content directly taken from pre-published sources is called plagiarized text. Text Comparison is needed to check unauthorized/ illegal usage of views, ideas & publications.So there is a need to find a suitable technique to find similarity between two documents.There are many text matching mechanisms such as Levenshtein’s Edit Distance, Cosine Similarity measure, Jaccard Similarity Coefficient, N gram, Hamming Distance, Scam Algorithm, Finger Printing, Substring Matching etc. But all these techniques have disadvantages like:

• Some techniques work on syntax whereas some of them are semantically sensitive.
• Many techniques fail due to the lack of computational resources and processing takes a large amount of time.

The chosen techniques are primitive, moderate and advanced respectively. So, the aim here is to enhance the aforesaid algorithms in terms of similarity index, time and to provide graphical comparison reports.

Internships closed for this project.

Principal Investigator:

Ms. Shama ME Scholar, CSE

Funding- Internal        Project Start Date - 03/04/2015

In India, IT Act 2000 has been established to combat the challenges of cyber crime which has become a global phenomenon. It must be well understood that victimization of women is on the higher side in comparison to men. There are several issues related to cyber crime against women which need to be thoroughly understood as it may have great impact on the security of women.

Even with the increase in the number of enrollment of women for higher education the women in India are still not open minded and shy in reporting crimes against them.  The threat of cyber abuses and cyber crimes is more dreadful because of the modus operandi of committing cyber crimes. It is not easy to understand the intention of the culprit. They can escape comfortably and can use different websites therefore to identify the existent of the culprit is more difficult in the case of cyber crimes.

There are various methods of providing cyber security to women in particular and society in general. But it can be possible through two ways. One woman needs to report all cyber crimes against them so that the types of threats can be understood clearly and second, the law needs to give proper security measures to women against cyber crimes.

Therefore the proposed research project is destined to achieve these twin objectives. One is to make women more aware regarding all possible cyber abuses and cyber crimes and informing them to use cyber security measures to keep the privacy of their information. Second is to suggest various improvement in IT Act 2000 to make it stronger and ensure victimized women that strong actions will be taken against the offender.

INTERNSHIPS ARE CLOSED FOR THIS PROJECT.

Principal Investigator:

Dr. Renuka Gupta, Dr. Kiran Mehta, Dr. K.S. Bath

Funding- External        Project Start Date - 01/04/2016

The global market economy is changing from a sellers’ market to a buyers’ market. The emerging field of neuromarketing, which makes use of brain research in a managerial context, has gained increasing popularity in the academic literature as well as the practical world. Neuromarketing is the branch of neuroscience research that aims at studying the consumer behaviour through the brain’s instinctive processes and responses. Such information can further be applied to understand marketing trends, explaining consumer’s preferences, motivations, expectations, predicting human behaviour and evaluating successes or failures of advertising messages. The global market for neuromarketing technology reached $21.0 million in 2015. The market is expected to reach $50.3 million by 2021 from $22.0 million in 2016, increasing at a compound annual growth rate (CAGR) of 18.0% from 2016 to 2021. One very big mega trend in the global economy is that of Big Data or Big Analytics. According to a report published by The Data Warehousing Institute, sentiment analytics and trending imbued with big data analytics can contribute up to 24% improvement in sales and marketing. One very attractive opportunity in big data is linking analytics with neuromarketing. This fact has been duly supported by the Gartner’s hype cycle graph, where the anticipated the scope of neuromarketing tactics and big data is expected to last for next ten years from now. Noteworthy, India, on the other hand, is the seventh largest nation in size, second most populated country and the seventh largest economy in the world. Further, the government of India is promoting make in India ideology under which, five industrial corridor projects have been identified, planned and launched by the Government of India in the Union Budget of 2014-2015, to provide an impetus to industrialization and planned urbanization. In each of these corridors, manufacturing will be a key economic driver and these projects are seen as critical in raising the share of manufacturing in India’s Gross Domestic Product from the current levels of 15% to 16% to 25% by 2022. The Micro, Small and Medium Enterprises (MSMEs) sector assumes a pivotal role in driving the growth engine and to ensure epitomizing vitality. According to Fortune from Times Inc., nine out of ten startups fail because of reasons like failing to understand the market need, not choosing the right team, poor marketing, ignoring consumers and limited liquid assets etc. There is no direct platform available for MSMEs to evaluate their marketing strategies, product design etc. which could prove to be a substitute for extensive and expensive market surveys followed by traditional way of analysing the collected data. The quality of data collected through surveys, interviews and questionnaires entirely depends upon the technical expertise of the investigator whereas nontraditional methods of collecting user response like eye tracking, electroencephalogramy (EEG), speech, and facial action coding can not only provide quality data but can also lead to precise business conclusions and decisions. User sentiment analysis information is already being used by businesses, governments and individuals to make timely and profitable decisions. Neuromarketing makes use of techniques like Neuro-Metrics, Eye-Tracking, Bio-Metrics, Implicit Response Testing and Facial Action Coding to understand/record human reactions. Sentiments analysis if infused with neuromarketing techniques can further provide automated mining of attitudes, opinions, expressions and emotions fromtext, speech, emotion, neuron activity and other database sources for making profitable business decisions. Data provided by neuromarketing studies are pictures and numbers, statistics. Neuro‐imagingtechnology, through its technical specifications, is able to provide accurate data for analysis and a cost effective solution to study market trends, hence providing major support to any economy.

Number of intern positions available in the project: 2

Principal Investigator:

Dr Jaiteg Singh

Funding- Internal        Project Start Date - 01/11/2017

Broad domain – Mechanical Engineering, Electronics & Computer Science

Focused domain – Automation, Sensors, Actuators

The present invention discloses an autonomous wall climbing the machine to be used for painting purpose. The control of the machine is dual in the sense that it can be controlled either manually with the PLC control unit available on the ground or machine itself can take self-control of its movement also as it is fully autonomous. Sensors are integrated with the machine which not only avoids an obstacle on the wall such as a window, glass sheet, pipes, hollow space etc. but also calculates the information about the specific area to be painted. The counter embedded with the suction cup counts the number of times the suction cup is used for generating the negative pressure which gives an indication about suction cup replacement before it malfunctions. For the protection of the machine from being fallen from the tall building during its operation, an air bag control unit in conjugation with an acceleration sensor is also embedded with the machine. So, the invention is light weight, cost effective and highly efficient.

No. of intern positions available in the project – 3

Principal Investigator:

Dr K.R.Ramkumar

Funding- Internal        Project Start Date - 01/11/2017

Broad domain – Electronics and Computer Science Engineering

Focused domain – Arduino Board Programming, Raspberry Pi, FPGA, Networking

The present invention discloses a mobile phone which always remains traceable as it cannot be switched OFF by an unauthorized user. The same is achieved using a fingerprint sensor and a fingerprint scanner fixed in the power button itself and the moment when a phone is stolen and if it is already put in the theft alarm mode, the device starts comparing the fingerprints. In the case of mismatch in the fingerprint, the phone goes into sleep mode, where the content(s) /contact(s) are not shown to anybody else and the phone will send all contact details to a centralized server(s)/ alternate phone(s). Also, the burglar cannot switch OFF/ON the mobile, the location details will be sent to the nearest police station without the burglar’s knowledge. If the device is not found in the stipulated time then all content(s)/contact(s) are deleted automatically making backup available on the server. The mobile service provider(s) can be requested to give a dedicated server to store data in case of stolen mobile devices.

No. of intern positions available in the project – 2.

Principal Investigator:

Dr K.R. Ramkumar

Funding- Internal        Project Start Date - 01/11/2017

Broad domain – Computer Science and Engineering

Focused domain – Network Security, Key management, Abstract

The proposal works on Key management for Mobile Ad hoc networks by using Numerical methods. A complete framework that executes key issue and key revocation for a secure way of communication is proposed. The numerical methods have a high impact on distributed key management which is essential for Mobile Ad hoc Networks, in particular, the Newton’s, Lagrange’s, curve fitting and Chebyshev polynomials are more suitable. The proposal gives a complete security framework for Mobile Ad hoc Networks.

No. of intern positions available in the project – 

Principal Investigator:

Dr K.R. Ramkumar

Funding- Internal        Project Start Date - 01/11/2017

Broad domain – Computer Science and Engineering, Electronics Engineering

Focused domain – FPGA, Number TheoryAbstract

1. Number Theory Basic concepts

2. Security Standards and Algorithms

3. Polynomials and Key Management

4. Cryptanalysis

5. Introduction to Hardware implementation of Network Security algorithms

6. Introduction to FPGA

7. ASIC Design Flow

8. Designing of VHDL/Verilog Code for Network Security Algorithm

9. Simulation of Code on ISE Project Navigator (Xilinx Software)

10. Hardware implementation of code on Artix-7 FPGA kit

No. of intern positions available in the project – 3

Principal Investigator:

Dr K.R. Ramkumar

Funding- Internal        Project Start Date - 01/11/2017

Broad domain – Sensor development, Data Mining and Data Science, Real-time Processing of Data, Wireless sensor networks, Arduino Processor

Focused domain – Data science and electromagnetics

Access to wireless sensor network facility at IOTREK Technology, Delhi and labs of DTU Delhi

The target is to pilot the project in University campus for monitoring of underwater pipe networking with help of data science and sensors at different nodes. The sensor nodes are capable of sensing physical, chemical and microbiological properties of water. The graphene based sensors are energy efficient and front end section embedded with the sensor allows remote monitoring of sensing information.

No. of intern positions available in the project – 01 (Engineering)

Principal Investigator:

Dr Nitin Saluja

Funding- External        Project Start Date - 01/11/2017

Broad domain – Food Processing, microwave engineering, thermodynamics, data processing and material science engineering

Focused domain – Mechanical Engineering with Food Processing

Farmers suffer from huge loss of fruits/vegetables and hence bound to sell it at quite cheap prices. The preservation techniques like drying seek lots of efforts but the recent advancements in microwave engineering and the group working on the project enable energy efficient, environmentally friendly, highly cost effective and scalable machine.

No. of intern positions available in the project – 01 (Engineering) and 01 (Hotel management)

Principal Investigator:

Dr Nitin Saluja

Funding- Internal        Project Start Date - 01/11/2017

Broad domain –Computer Vision
Focused domain –Machine Learning

The predictive analysis using the machine learning has been incorporated to meet the requirements of growing era of machine learning and automation among the IT industry. In the coming projects, the work will be carried out towards the process automation, pattern recognition, data lineage, etc using the intelligent machine learning mechanisms. The main motive behind the consideration upon the research and innovation in the IT industry automation is to meet the standards and regulations for the implementation of the real-time projects. The projects would be carried forward with the enrollment in a competition held by “KAGGLE”, which is a popular for offering the competitions in the field of predictive modelling and related stuff for analysts, researchers, statisticians and data miners to meet up the new challenges of the industry. Various real time data sets are offered with the well-defined problems, mostly with prizes on the stake, undergo the KAGGLE data science competitive space. The enrollment for the competitions in predictive modelling and industrial automation (IT) on KAGGLE is most likely for the following competitions:
1. Passenger Screening Algorithm Challenge
2. Zillow Price: Zillow’s Home Value Prediction (Zestimate)
3. Draper Satellite Image Chronology
4. CIFAR-10
5. Depth Generation – Lightfield Imaging

ENROLLMENT OF STUDENTS
The enrollment of the students can be considered for the various roles with different responsibilities during the development of the model according to the life cycle of development of the projects. The following roles can be offered to the students during the project’s life cycle:
1. Literature study assistance
2. Programmers (Python, R, Scala, etc)
3. Data Visualization
4. Data preparation
5. Data analysis & report generation
6. Model’s performance evaluation
7. Segmental Analysis approach (Report generation and data integration/segmentation)
8. Reporting, Documentation and Packaging

No. of intern positions available: 02

Principal Investigator:

Dr Sheifali Gupta

Funding- External        Project Start Date - 01/11/2017

Broad domain – Image Processing
Focused domain – Image Segmentation

Carvana, a successful online used car startup, has seen an opportunity to build long term trust with consumers and streamline the online buying process.An interesting part of their innovation is a custom rotating photo studio that automatically captures and processes 16 standard images of each vehicle in their inventory. While Carvana takes
high-quality photos, bright reflections and cars with similar colours as the background cause automation errors, which requires a skilled photo editor to change.

In this competition, you’re challenged to develop an algorithm that automatically removes the photo studio background. This will allow Carvana to superimpose cars on a variety of backgrounds. You’ll be analysing a dataset of photos, covering different vehicles with a wide variety of year, make, and model combinations.

Prize Money

1st place – $12,000
2nd place – $8,000
3rd place – $5,000

ENROLLMENT OF STUDENTS

The enrollment of the students can be considered for the various roles with different responsibilities during the development of the model. The students must have following pre-requisites for the enrollment
 Programming Skills
 Working Knowledge in Python
 MATLAB Basics

Number of Positions: 02

Principal Investigator:

Dr Sheifali Gupta

Funding- External        Project Start Date - 01/11/2017

Broad domain – Automotive engineering
Focused domain – Vehicle dynamics and powertrain systems

IMechE`s(Institution of Mechanical Engineers) Formula Student is an annually organised student motorsport competition in which students are given a task to design an open wheel, single seater Formula 1 style race-car under certain rules described by IMechE. Each year, the student teams gather at Silverstone racing circuit in the month of July and teams are judged on the basis of Cost report, design presentations, business presentation and dynamic events like acceleration, endurance, fuel economy etc. MNCs like Ricardo, Mercedes- AMG, Jaguar Land Rover recruit students from this competition. For the next season, the team is going to participate in Formula Bharat also which is going to be held from Jan 25- Jan 28, 2018, at Kari Motor Speedway in Coimbatore. For more info, please visit Formula Student Website.

Students required- Mechanical engineers- 3 (powertrain systems and Vehicle dynamics)

Electronics and communication engineers- 1 (For engine electronics, car information system and DA systems)

Business administration- 1 (preferably girl for marketing, business presentation and branding)

Give away to students- Automobile designing experience+ certificates + chance to be placed in top auto-manufactures( for business administration aspirants also).

Principal Investigator:

Varinder Singh

Funding- Internal and External        Project Start Date - 01/11/2017