Effect of the regularization parameter on image-guided filter
The guided filter is derived from a local linear model. It computes the filtering output by considering the content of a guidance image (self-image or reference image). The guided filter can be used as an edge-preserving smoothing operator that behaves better near edges. It can transfer the structures of the guidance image to the filtering output, enabling new filtering applications like dehazing and guided feathering. Moreover, the guided filter naturally has a fast and non-approximate linear time algorithm, regardless of the kernel size and the intensity range. Regularization parameter (ϵ) is one of the two important factors which determines the characteristics of the guided image applied to the input image i.e. it helps to perceive the degree of the edges of an image. In this paper, we will observe the effect of the regularization parameter on a guided filtered image.
Published by: Eisita Basak, Sangita Roy, Dr. Saradindu Panda
Author: Eisita Basak
Paper ID: V6I3-1539
Paper Status: published
Published: June 20, 2020
Privacy policies of health care companies
The need for and importance of privacy policies across the healthcare continuum is based on the basic and fundamental human rights. International laws and regulations are linked to the physician’s oath of confidentiality and secrecy. The general medical council provides formwork for healthcare to design the privacy policy to protect confidentiality. Various fields across healthcare continue like a chronic disease which might make a patient terminally ill for e.g. Human immune deficiency virus (HIV), types of cancers make patients vulnerable and depressed to a great dept. Humans are social animals and are psychologically affected by treatment. Studies are shown that the placebo effect at patients who are terminally ill has given them a chance to be more psychologically fir and healthier. The second example is international surrogacy laws, confidentiality, and privacy policies are of utmost legal importance determining the paternity of the child. Confidentiality and privacy policies are made and communicated to all stakeholders including staff and patients to ensure correct implementation and full protection of patient’s data. According to the federal of the country, every organization needs to maintain privacy in policies. In an age where technology is overpowering the healthcare industry and cybersecurity threat is always looming above the head, there is a need to develop and implement more stringent privacy policies. Confidentiality and privacy policies must be communicated to the end-users so as to ensure their consent for the treatment and to boost their confidence, their data, and their treatment. These data must remain confidential at all costs. Confidentiality must be maintained while it is processing. The data processing involves creating, storing, and sharing.
Published by: Akshay Mittal
Author: Akshay Mittal
Paper ID: V6I3-1553
Paper Status: published
Published: June 18, 2020
IoT based air and sound pollution monitoring system
The transportation sector is the major contributor to increasing air pollution levels in major cities in India. Various authorities like Road and Transportation Office monitor emission levels. This paper states how using various sensors we can measure emission levels. The data collected is compared to the standards. If it doesn’t meet the standards, the vehicle owner receives an alert and the data is also sent to the RTO for further actions to be taken if needed. Thus monitoring the exhaust for presence and levels of harmful gases emitted, pollution levels can significantly be reduced.
Published by: Vidyashree S., Shalini S. V., Shilpa M. Udikeri, Shreyas K. R.
Author: Vidyashree S.
Paper ID: V6I3-1556
Paper Status: published
Published: June 18, 2020
Student placement possibility prediction using Naive Bayes algorithm
The aim of my project is to analyze how we can use available student data to predict the standard of new students and the possibility of their placement. It’s a student assessment on the basis of his academic skills, Innovation, Research, and Development capabilities. Doing this can help students as well as academics to boost students' standards. Also, it helps us to design our course and extra training on the basis of it for maximum benefits to students. For this, we conducted a small survey using Google form survey and collected information about students of different colleges and universities around Chhattisgarh in the Computer science & Engineering Department and Information Technology Department. Creating a dataset for this project is not easy because motivating students to share their information is a challenging task. For predicting the result of student placement possibility we use the dataset prepared from the survey and different types of Naïve Bayes Algorithm, which is a Supervised Machine Learning Algorithm. Naïve Bayes Classifier is one of the most effective and simplest algorithms to implement for a supervised labeled dataset.
Published by: Brijmohan Lal Sahu, Dr. Anil Tiwari
Author: Brijmohan Lal Sahu
Paper ID: V6I3-1555
Paper Status: published
Published: June 18, 2020
Fashion recommendation system using CNN
Recommendation systems are the techniques that are used to predict the rating one individual will give to an item or social entity. The items can include books, movies, restaurants, and things on which individuals have different preferences. These preferences are being predicted using two approaches first content-based approach which involves characteristics of an item and second collaborative filtering approaches which considers the user's past behavior to evaluate its choices. This thesis proposes a fashion recommendation system that will recommend clothing images supported the style sort of the provided clothing images. In this work, we focus on the images of the upper body as well as the lower body clothing and with the human model in the images. We have created our own datasets through web scrapping of different e-commerce websites. In this paper, we have come up with an idea to build a content-based recommendation system using ResNet-50 convolutional neural network.
Published by: Anjan M., Abhishek V., C. Balamanikantan, Dheeraj, Dr. Venugeetha Y.
Author: Anjan M.
Paper ID: V6I3-1546
Paper Status: published
Published: June 18, 2020
An exploratory study on the perspectives and practices of contraceptive use among married females of reproductive age group (15-49 years) of selected urban areas of District Sirmour (Himachal Pradesh)
Contraception is a process or technique for preventing pregnancy by means of medication, device or method that blocks or alters one or more of the processes of reproduction in such a way that sexual union can occur without impregnantion. Objectives: To assess the perspectives of contraceptive use among married females of selected urban areas of Dist. Sirmour (HP), to identify the contraceptive practices among married females of selected urban areas of Dist. Sirmour (HP), to associate the perspectives and practices of contraceptive use with the selected socio demographic variables of married females. Method: The study has adopted quantitative research approach and exploratory research design. Total 238 samples were selected based on inclusion and exclusion criteria through multistage cluster sampling technique. Data was collected in terms of socio demographic profile of the married females, semi structured questionnaire and likert scale was used to assess the perspectives of contraceptives use, semi structured checklist was used to assess the practices of contraceptives among married females. Results: Major findings of the study revealed that most of the married females i.e. 111(47%) were between the age group of 20-29 years, 171(72%) belonged to hindu religion, 113(47%) were graduate or above, 75(32%) were self-employed, 137(58%) were having duration of marriage between 1-10 years, 203(85%) were living with spouse and children, 152(64%) were having 1 child, 94(39%) had monthly family income (in rupees) between 20001-30000, 157(66%) married females were living in nuclear family, 216(91%) married females had television as the source of information and contraceptive method used was condom in 169(71%) married females. In the present study, the perspectives of married females in terms of awareness showed that 35% of females were having good awareness, 50% were having average awareness and 15% were having poor awareness. In terms of opinion, 30% of married females were having positive opinion, 70% were having neutral opinion and none of them was having negative opinion regarding contraceptive use. Majority (65%) of married females having average practices of contraceptives, only 31% of females were having poor practices and the rest 4% were having good practices. Mean and standard deviation for perspectives of contraceptive use in relation to awareness and opinion was 14.66 ± 3.98, 72.16 ± 8.24. Mean and standard deviation for practices of contraceptives were 7.71 ± 2.42. There was significant association of age, education, occupation, duration of marriage, living status, monthly family income(in rupees), type of family, source of information and contraceptive method used with perspectives of contraceptive use at p<0.05 level of significance. There was significant association of education, occupation, duration of marriage, monthly family income( in rupees), type of family, source of information and contraceptive method used with practices of contraceptives at p< 0.05 level of significance.
Published by: Reena Chaudhary, Anna Bajaj
Author: Reena Chaudhary
Paper ID: V6I3-1530
Paper Status: published
Published: June 18, 2020
