Pub Date : 2020-12-04DOI: 10.1109/IBSSC51096.2020.9332183
Zeeshan I. Khan, V. Shandilya
This paper presents an authentication scheme that contains images, alphabets, numbers and symbols. The primary aim of this research is to develop the authentication scheme using images as the front end data which makes it easy to remember for the user. But to store that selected images by the user in the backend (database), it takes a large amount of memory as well as processing time for password verification. To prevent this, the proposed method converts all the selected images in the string containing alphabets, numbers & symbols. This method focuses on recognition based image authentication scheme and also provides its two different ways of authentication. Sequential recognition based and Non-Sequential recognition-based image authentication schemes which will provide a facility to the user to verify its password sequentially or non-sequentially. At the user side, this authentication scheme is flexible & easy to remember. And at the system side, it provides low memory consumption and verifies the password in less amount of time. After the scheme is developed, the system time & memory is also calculated and analyzed. It has been observed that this scheme is saving the time and memory making the system advantageous.
{"title":"Enhanced Recognition Based Image Authentication Scheme to Save System Time & Memory","authors":"Zeeshan I. Khan, V. Shandilya","doi":"10.1109/IBSSC51096.2020.9332183","DOIUrl":"https://doi.org/10.1109/IBSSC51096.2020.9332183","url":null,"abstract":"This paper presents an authentication scheme that contains images, alphabets, numbers and symbols. The primary aim of this research is to develop the authentication scheme using images as the front end data which makes it easy to remember for the user. But to store that selected images by the user in the backend (database), it takes a large amount of memory as well as processing time for password verification. To prevent this, the proposed method converts all the selected images in the string containing alphabets, numbers & symbols. This method focuses on recognition based image authentication scheme and also provides its two different ways of authentication. Sequential recognition based and Non-Sequential recognition-based image authentication schemes which will provide a facility to the user to verify its password sequentially or non-sequentially. At the user side, this authentication scheme is flexible & easy to remember. And at the system side, it provides low memory consumption and verifies the password in less amount of time. After the scheme is developed, the system time & memory is also calculated and analyzed. It has been observed that this scheme is saving the time and memory making the system advantageous.","PeriodicalId":432093,"journal":{"name":"2020 IEEE Bombay Section Signature Conference (IBSSC)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132910109","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-12-04DOI: 10.1109/IBSSC51096.2020.9332178
Sonaali Borkar, Meenakshi Nandula
Conventional classroom teaching has been continually evolving, has undergone several technological onslaughts. The global pandemic crisis has enforced the shift of conventional classroom teaching to online classrooms. The sudden advent of online teaching has created a plethora of challenges in learning and has thus put in the same repercussions on teaching. A survey of teachers undergoing this shift was conducted to understand the current experience of online teaching to recognize the differences between classroom teaching and online teaching approaches, and identify the challenges therein. SWOT analysis of the teachers is presented in this paper to understand Strengths, Weaknesses, Opportunities and Threats. The study reveals that teachers have started teaching students online even without adequate training. This paper attempts to identify the crux of the shift to the online method of teaching. This would not only differentiate between the two methodologies, but bridge the gap in the missing nonverbal and face to face communication in online methodology. The opinions of expert educators also add an additional dimension to the SWOT analysis. A few measures based on observations and experience as teachers are suggested, and further open this topic for brainstorming to make the experience of teaching-learning a more impactful one for teachers and students.
{"title":"The paradigm shift towards e-Teaching: SWOT analysis from the perspective of Indian teachers","authors":"Sonaali Borkar, Meenakshi Nandula","doi":"10.1109/IBSSC51096.2020.9332178","DOIUrl":"https://doi.org/10.1109/IBSSC51096.2020.9332178","url":null,"abstract":"Conventional classroom teaching has been continually evolving, has undergone several technological onslaughts. The global pandemic crisis has enforced the shift of conventional classroom teaching to online classrooms. The sudden advent of online teaching has created a plethora of challenges in learning and has thus put in the same repercussions on teaching. A survey of teachers undergoing this shift was conducted to understand the current experience of online teaching to recognize the differences between classroom teaching and online teaching approaches, and identify the challenges therein. SWOT analysis of the teachers is presented in this paper to understand Strengths, Weaknesses, Opportunities and Threats. The study reveals that teachers have started teaching students online even without adequate training. This paper attempts to identify the crux of the shift to the online method of teaching. This would not only differentiate between the two methodologies, but bridge the gap in the missing nonverbal and face to face communication in online methodology. The opinions of expert educators also add an additional dimension to the SWOT analysis. A few measures based on observations and experience as teachers are suggested, and further open this topic for brainstorming to make the experience of teaching-learning a more impactful one for teachers and students.","PeriodicalId":432093,"journal":{"name":"2020 IEEE Bombay Section Signature Conference (IBSSC)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114379604","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-12-04DOI: 10.1109/IBSSC51096.2020.9332176
Anjali Mishra, S. Behra
The concept of Aerial Manipulation refers to the design of Unmanned Aerial Vehicle to work as a combined system in collaboration with an object grasping device. Over the past two decades there have been tremendous discoveries regarding the different methodologies used for aerial manipulation which yield the best results in terms of flight time, stability, object grasping and releasing. The Quadcopter is intended to serve as an aid in Search and Rescue Operations where the Drone’s presence will function as a better alternative to human resources. The focus of the proposed work is to enable the Drone to have a simple claw type grasping mechanism which is specially designed keeping in mind important designing specifications of the UAV such as stability, payload lifting capability etc. which will be single handedly controlled via the Flight Controller which is an ARM Cortex M4 based Microcontroller and with the help of an open source Autopilot powered GUI to set the navigation points via the telemetry module.
{"title":"Design of an Aerial Manipulation System with Robotic Claw for Search and Rescue Operations","authors":"Anjali Mishra, S. Behra","doi":"10.1109/IBSSC51096.2020.9332176","DOIUrl":"https://doi.org/10.1109/IBSSC51096.2020.9332176","url":null,"abstract":"The concept of Aerial Manipulation refers to the design of Unmanned Aerial Vehicle to work as a combined system in collaboration with an object grasping device. Over the past two decades there have been tremendous discoveries regarding the different methodologies used for aerial manipulation which yield the best results in terms of flight time, stability, object grasping and releasing. The Quadcopter is intended to serve as an aid in Search and Rescue Operations where the Drone’s presence will function as a better alternative to human resources. The focus of the proposed work is to enable the Drone to have a simple claw type grasping mechanism which is specially designed keeping in mind important designing specifications of the UAV such as stability, payload lifting capability etc. which will be single handedly controlled via the Flight Controller which is an ARM Cortex M4 based Microcontroller and with the help of an open source Autopilot powered GUI to set the navigation points via the telemetry module.","PeriodicalId":432093,"journal":{"name":"2020 IEEE Bombay Section Signature Conference (IBSSC)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123747834","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-12-04DOI: 10.1109/IBSSC51096.2020.9332172
M. A. Shaikh, Sneha Annappanavar
The use of Online Clickbait in different social media platforms have increased momentarily. Basically, click baits are the eye-catching titles or headlines which exaggerate the facts and make the user to “click” on it. These clickbaits comes in many forms like images, videos also through advertisements. This links will lead you to anonymous websites which contains very little information and create nuisance on the internet. In social media clickbaits are very commonly used and Detection of Clickbait is a very crucial process. This paper proposes a method using a deep learning algorithm namely Convolution Neural Network (CNN) for detecting the clickbaits on the social media platforms. The used method focuses on the textual features which consider the word sequence information and also learns the word meanings from entire dataset. Our Results obtained a high accuracy of 0.82% comparatively better than different Machine Learning algorithms. We also did comparative analysis with the classification algorithm called Random Forest (RF).
{"title":"A Comparative Approach For Clickbait Detection Using Deep Learning","authors":"M. A. Shaikh, Sneha Annappanavar","doi":"10.1109/IBSSC51096.2020.9332172","DOIUrl":"https://doi.org/10.1109/IBSSC51096.2020.9332172","url":null,"abstract":"The use of Online Clickbait in different social media platforms have increased momentarily. Basically, click baits are the eye-catching titles or headlines which exaggerate the facts and make the user to “click” on it. These clickbaits comes in many forms like images, videos also through advertisements. This links will lead you to anonymous websites which contains very little information and create nuisance on the internet. In social media clickbaits are very commonly used and Detection of Clickbait is a very crucial process. This paper proposes a method using a deep learning algorithm namely Convolution Neural Network (CNN) for detecting the clickbaits on the social media platforms. The used method focuses on the textual features which consider the word sequence information and also learns the word meanings from entire dataset. Our Results obtained a high accuracy of 0.82% comparatively better than different Machine Learning algorithms. We also did comparative analysis with the classification algorithm called Random Forest (RF).","PeriodicalId":432093,"journal":{"name":"2020 IEEE Bombay Section Signature Conference (IBSSC)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125389451","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-12-04DOI: 10.1109/IBSSC51096.2020.9332222
Ujwala Bharambe, Chhaya Narvekar, Siddhesh Shinde
Examination performance is the only measure of competence in the existing education system of India. Question paper is the primary tool used in an examination, the quality of question paper plays an important role in a student’s future. Hence, any individual should take utmost care while framing the question paper. However, setting up a good question paper for assessment is not a straightforward task, particularly when students come from different backgrounds and intellect. So parameters which need to pay attention are fairness, consistency, novelty and elimination of bias while selecting questions in the question paper. This paper explores the possibility of using knowledge graph technology and deep learning. Knowledge graph in particular, addressing syllabus fairness and deep learning for novelty check. The objective is to propose a feasible framework, taking into consideration several aspects. These aspects are: to meet at least three levels of bloom taxonomy, support of course outcomes (CO) and Program Outcomes (PO) attainment. They are achieved by judging syllabus fairness, support for judging more spontaneous answering potential with novelty checks. The prototype implemented using this framework showed promising results motivating for further research.
{"title":"Fairness Assessment of Question Paper using Artificial Intelligent Techniques","authors":"Ujwala Bharambe, Chhaya Narvekar, Siddhesh Shinde","doi":"10.1109/IBSSC51096.2020.9332222","DOIUrl":"https://doi.org/10.1109/IBSSC51096.2020.9332222","url":null,"abstract":"Examination performance is the only measure of competence in the existing education system of India. Question paper is the primary tool used in an examination, the quality of question paper plays an important role in a student’s future. Hence, any individual should take utmost care while framing the question paper. However, setting up a good question paper for assessment is not a straightforward task, particularly when students come from different backgrounds and intellect. So parameters which need to pay attention are fairness, consistency, novelty and elimination of bias while selecting questions in the question paper. This paper explores the possibility of using knowledge graph technology and deep learning. Knowledge graph in particular, addressing syllabus fairness and deep learning for novelty check. The objective is to propose a feasible framework, taking into consideration several aspects. These aspects are: to meet at least three levels of bloom taxonomy, support of course outcomes (CO) and Program Outcomes (PO) attainment. They are achieved by judging syllabus fairness, support for judging more spontaneous answering potential with novelty checks. The prototype implemented using this framework showed promising results motivating for further research.","PeriodicalId":432093,"journal":{"name":"2020 IEEE Bombay Section Signature Conference (IBSSC)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127428711","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-12-04DOI: 10.1109/IBSSC51096.2020.9332177
O. Oteri
Most technical courses mainly in engineering, science and technology, have always been taught using very expensive equipment usually stationed in a particular lab setup in the university with students accessing these labs in groups. The problem is even compounded further in the case where the number of students is high in a class. This paper discusses the use of a mobile lab being a simple, inexpensive and portable kit to undertake practicals in the fields of engineering, science and technology a case study of digital electronics, computer organization, networking and operating systems that are done by students in these fields. The study involved the development of IoT layer one based kit in terms of hardware and software support together with its actual use in a classroom setup. For full implementation, a laptop or a smartphone are also required. The first group of students, who were requested to do an evaluation of the kit at the end of the semester, was undertaking computer science. The results were quite positive showing the link between IoT and a classroom setup making it easy for the students to relate what they learn in class and the many IoT related actual real life applications.
{"title":"The Application of IoT layer one Based Mobile Labs in Engineering, Science and Technology Education","authors":"O. Oteri","doi":"10.1109/IBSSC51096.2020.9332177","DOIUrl":"https://doi.org/10.1109/IBSSC51096.2020.9332177","url":null,"abstract":"Most technical courses mainly in engineering, science and technology, have always been taught using very expensive equipment usually stationed in a particular lab setup in the university with students accessing these labs in groups. The problem is even compounded further in the case where the number of students is high in a class. This paper discusses the use of a mobile lab being a simple, inexpensive and portable kit to undertake practicals in the fields of engineering, science and technology a case study of digital electronics, computer organization, networking and operating systems that are done by students in these fields. The study involved the development of IoT layer one based kit in terms of hardware and software support together with its actual use in a classroom setup. For full implementation, a laptop or a smartphone are also required. The first group of students, who were requested to do an evaluation of the kit at the end of the semester, was undertaking computer science. The results were quite positive showing the link between IoT and a classroom setup making it easy for the students to relate what they learn in class and the many IoT related actual real life applications.","PeriodicalId":432093,"journal":{"name":"2020 IEEE Bombay Section Signature Conference (IBSSC)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116907792","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-12-04DOI: 10.1109/IBSSC51096.2020.9332210
R. Bhadra, Subhajit Kar
Severe Acute Respiratory Syndrome Corona virus 2 (SARS-COV-2) also known as COVID-19 has been emerged as a pandemic throughout the globe recently. Therefore, accurate diagnosis of COVID-19 is necessary to fight against this pandemic situation. In this context, chest X-ray (CXR) scans play an important role in the diagnosis of the corona virus. In this paper, an intelligent detection and classification technique of COVID-19 has been proposed to assist doctors in their diagnostic prediction. A deep multi-layered convolution neural network (CNN) has been proposed to detect COVID-19 accurately from CXR scans. The proposed methodology has experimented on a combination of multiple open source publicly available datasets. Experimental results demonstrate the efficacy of the proposed methodology in COVID-19 detection from CXR images.
{"title":"Covid Detection from CXR Scans using Deep Multi-layered CNN","authors":"R. Bhadra, Subhajit Kar","doi":"10.1109/IBSSC51096.2020.9332210","DOIUrl":"https://doi.org/10.1109/IBSSC51096.2020.9332210","url":null,"abstract":"Severe Acute Respiratory Syndrome Corona virus 2 (SARS-COV-2) also known as COVID-19 has been emerged as a pandemic throughout the globe recently. Therefore, accurate diagnosis of COVID-19 is necessary to fight against this pandemic situation. In this context, chest X-ray (CXR) scans play an important role in the diagnosis of the corona virus. In this paper, an intelligent detection and classification technique of COVID-19 has been proposed to assist doctors in their diagnostic prediction. A deep multi-layered convolution neural network (CNN) has been proposed to detect COVID-19 accurately from CXR scans. The proposed methodology has experimented on a combination of multiple open source publicly available datasets. Experimental results demonstrate the efficacy of the proposed methodology in COVID-19 detection from CXR images.","PeriodicalId":432093,"journal":{"name":"2020 IEEE Bombay Section Signature Conference (IBSSC)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133010163","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-12-04DOI: 10.1109/IBSSC51096.2020.9332213
B. N. K. Reddy, G. Reddy, B. Vani
Pseudorandom Number Generator (PRNG) are used in Built in Self Test (BIST) to reduce testing cost and time. The linear feedback shift register (LFSR) pattern generator is mostly used in generating test vectors for PRNG. LFSRs play a vital role in generating test vectors in hardware verification or testing and they are also employed in the cryptography area. This paper presents the design of 4-bit LFSR with 2 Phase clocked Adiabatic Static CMOS Logic (2-PASCL) and Reversible Logic Gates (RLG). The proposed 4-bit LFSR is synthesized and simulated using Vivado design suit 2018.3 and implemented on a Kintex-7 FPGA board. Compared with the results obtained with well-known LFSR architectures, the proposed method is used to improve the performance, and decrease the power consumption and area of processors.
{"title":"Design and Implementation of an Efficient LFSR using 2-PASCL and Reversible Logic Gates","authors":"B. N. K. Reddy, G. Reddy, B. Vani","doi":"10.1109/IBSSC51096.2020.9332213","DOIUrl":"https://doi.org/10.1109/IBSSC51096.2020.9332213","url":null,"abstract":"Pseudorandom Number Generator (PRNG) are used in Built in Self Test (BIST) to reduce testing cost and time. The linear feedback shift register (LFSR) pattern generator is mostly used in generating test vectors for PRNG. LFSRs play a vital role in generating test vectors in hardware verification or testing and they are also employed in the cryptography area. This paper presents the design of 4-bit LFSR with 2 Phase clocked Adiabatic Static CMOS Logic (2-PASCL) and Reversible Logic Gates (RLG). The proposed 4-bit LFSR is synthesized and simulated using Vivado design suit 2018.3 and implemented on a Kintex-7 FPGA board. Compared with the results obtained with well-known LFSR architectures, the proposed method is used to improve the performance, and decrease the power consumption and area of processors.","PeriodicalId":432093,"journal":{"name":"2020 IEEE Bombay Section Signature Conference (IBSSC)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124190219","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-12-04DOI: 10.1109/IBSSC51096.2020.9332179
Ms. Kimberly Morais
We are in the internet age but teaching power electronics is still done in the classical way. In COVID times we are discussing a lot about e-learning which needs investment. The amount of investment needed can be reduced with use of free/open source software. The idea of teaching power electronics with the aid of open source simulation software eSim will be discussed here. eSim is an open source EDA tool can be used for lectures, conduction of practicals and self- study by the students.
{"title":"Teaching Power Electronics with the Aid of Open Source Simulation Tool eSim","authors":"Ms. Kimberly Morais","doi":"10.1109/IBSSC51096.2020.9332179","DOIUrl":"https://doi.org/10.1109/IBSSC51096.2020.9332179","url":null,"abstract":"We are in the internet age but teaching power electronics is still done in the classical way. In COVID times we are discussing a lot about e-learning which needs investment. The amount of investment needed can be reduced with use of free/open source software. The idea of teaching power electronics with the aid of open source simulation software eSim will be discussed here. eSim is an open source EDA tool can be used for lectures, conduction of practicals and self- study by the students.","PeriodicalId":432093,"journal":{"name":"2020 IEEE Bombay Section Signature Conference (IBSSC)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115856151","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-12-04DOI: 10.1109/IBSSC51096.2020.9332175
C. Bhuma, Ramanjaneyulu Kongara
In this work, a deep learning methodology for accurate classification of histological images of the patients suffering from childhood medulloblastoma is proposed. Pre trained EfficientNets trained on the ImageNet dataset are considered in this work. Features are extracted from the average pooling layer of EfficienNets and are given to an error correcting output code classifier. Ensemble prediction from the selected pre-trained EfficientNets is employed. For the multi class classification, the proposed approach is able to predict with a mean classification accuracy of 98.78% for 10x level images and 95.67% for the 100x level images for an 80% train and 20% test split. The peak classification accuracy is 100% for both binary and multiclass case at cell level and architectural level. For the binary classification with same split, 100% mean classification accuracy is achieved even without ensemble prediction. The results are compared with an existing work on a similar dataset and the significant improvement is demonstrated with the experimental simulations.
{"title":"Childhood Medulloblastoma Classification Using EfficientNets","authors":"C. Bhuma, Ramanjaneyulu Kongara","doi":"10.1109/IBSSC51096.2020.9332175","DOIUrl":"https://doi.org/10.1109/IBSSC51096.2020.9332175","url":null,"abstract":"In this work, a deep learning methodology for accurate classification of histological images of the patients suffering from childhood medulloblastoma is proposed. Pre trained EfficientNets trained on the ImageNet dataset are considered in this work. Features are extracted from the average pooling layer of EfficienNets and are given to an error correcting output code classifier. Ensemble prediction from the selected pre-trained EfficientNets is employed. For the multi class classification, the proposed approach is able to predict with a mean classification accuracy of 98.78% for 10x level images and 95.67% for the 100x level images for an 80% train and 20% test split. The peak classification accuracy is 100% for both binary and multiclass case at cell level and architectural level. For the binary classification with same split, 100% mean classification accuracy is achieved even without ensemble prediction. The results are compared with an existing work on a similar dataset and the significant improvement is demonstrated with the experimental simulations.","PeriodicalId":432093,"journal":{"name":"2020 IEEE Bombay Section Signature Conference (IBSSC)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114527706","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：