Lakshmi Shrinivasan - Academia.edu (original) (raw)

Papers by Lakshmi Shrinivasan

Glaucoma is a disease that affects the optic nerve. This disease, over a period of time, can lead... more Glaucoma is a disease that affects the optic nerve. This disease, over a period of time, can lead to loss of vision. It is also known as 'silent thief of sight'. This is because this disease slowly damages the eye, and ultimately causes irreparable harm before any vision loss. There are several methods in which the disease can be treated, if detected at an early stage. It is definitely not possible for any technology, including artificial intelligence, to replace a doctor. However, it is possible to develop a model based on several classical image processing algorithms, combined with artificial intelligence that can detect onset of Glaucoma based on certain parameters of the retinal fundus. This model would play an important role in early detection of the disease and assist the doctor. The traditional methods to detect glaucoma, as efficient as they may be, are usually expensive. Here we propose a machine learning approach to diagnose from fundus images and accurately classify its severity. In this paper we propose Support Vector Machine (SVM) method to segregate, train the models using a high-end graphics processor unit (GPU) and augmented the hull convex approach to boost the accuracy of the image processing mechanisms along with distinguishing the different stages of glaucoma. Added to these, we have proposed a feasible web application for the screening process.

CRC Press eBooks, Feb 9, 2024

2022 International Conference on Industry 4.0 Technology (I4Tech)

2022 IEEE 4th International Conference on Cybernetics, Cognition and Machine Learning Applications (ICCCMLA)

IAES International Journal of Artificial Intelligence (IJ-AI)

Glaucoma is a disease that affects the optic nerve. This disease, over a period of time, can lead... more Glaucoma is a disease that affects the optic nerve. This disease, over a period of time, can lead to loss of vision. Which is known as ‘silent thief of sight’. There are several methods in which the disease can be treated, if detected at an early stage It is not possible for any technology, including artificial intelligence, to replace a doctor. However, it is possible to develop a model based on several classical image processing algorithms, combined with artificial intelligence that can detect onset of glaucoma based on certain parameters of the retinal fundus. This model would play an important role in early detection of the disease and assist the doctor. The traditional methods to detect glaucoma, as efficient as they may be, are usually expensive, a machine learning approach to diagnose from fundus images and accurately classify its severity can be considered to be efficient. Here we propose support vector machine (SVM) method to segregate, train the models using a high-end gra...

2022 IEEE 4th International Conference on Cybernetics, Cognition and Machine Learning Applications (ICCCMLA)

Journal of Engineering

The aim of this article is to address the problem of protecting the private property of a protect... more The aim of this article is to address the problem of protecting the private property of a protected object, namely: we propose an algorithm for detection of object movements by means of a neural network for the video surveillance system. Consistency of perception of the external world in the form of images allows for the investigation of properties of the limited number of objects on the basis of familiarization with their final number. Based on the literature analysis, the main definitions of the theory of image recognition were established, such as “image,” “sign,” and “vector realization.” A comparison is made of approaches, methods, and technologies for recognizing the movement of objects, and their strengths and weaknesses are discussed. It was found that the neuron network is the most effective method for solving the problem of recognition of the movement of objects due to the accuracy of the result, simplicity, and speed. On the basis of the structural scheme of the complex a...

Solid State Technology, Feb 22, 2021

Autonomous navigation is becoming a norm in all kinds of transport system and robotics, and the p... more Autonomous navigation is becoming a norm in all kinds of transport system and robotics, and the primary component of such system is the proximity sensors or rangefinders for creating the localized map of the environment they are in to avoid obstacles. The processing and computational power required to perform such an in the current technology is too high to be feasible for every day applications or unmanned aerial vehicles. In this report, we present an affordable low resource consumption 3 dimensional range sensor that provides a comparable performance with commercial grade range sensors which is easy to implement and scalable.

Emerging Research in Computing, Information, Communication and Applications, 2019

This research is a solution to the problem of a hardware platform of object tracking drones. This... more This research is a solution to the problem of a hardware platform of object tracking drones. This platform can be used to build further advancements such as selfie drones, follow-me drones for adventure sports and robot pets. StaGaus algorithm has been derived using first principles and is compared against standard algorithms like SRDCF. The algorithm is tested on, an on-board two Android phones for real-time telemetry. This paper demonstrates that StaGaus works even on memory- and performance-constrained devices. In order to standardise the development of object tracking drones’ algorithms, we have built a customised ROS-based Gazebo simulator from scratch. This simulator is capable of simulating multiple robots of multiple types. This uses actual physics Open Dynamics Engine which has been compared against the existing simulators. Finally, as a by-product, this design proved that the cost of the proposed physical quadcopter is low.

Lecture Notes in Electrical Engineering, 2016

Automatic modulation recognition is the vital part in the advanced communication system used for ... more Automatic modulation recognition is the vital part in the advanced communication system used for both military and civil applications. In this paper a new methodology is proposed for distinguishing five digital modulation schemes (ASK-2, ASK-4, FSK, BPSK and QPSK). The algorithm extracts the features from the received signal and they are tested against preset thresholds to determine the modulation type of received signal. The simulations are done using MATLAB 2013 and results show that the system has an average recognition rate of 99.6 % at SNR as low as 4 dB.

Procedia Computer Science, 2016

In a hybrid road network with multiple paths to same location having prior geographical knowledge... more In a hybrid road network with multiple paths to same location having prior geographical knowledge, successful navigation for mobile robots is one of the main challenges. Path planning is one of the most important issues in the navigation process which enables the selection and identification of a suitable path for the robot to traverse in the workspace area. Path-planning for mapped roads can be considered as the process of navigating a mobile robot around a configured road map, which provides optimized path by considering roughness of roads. In this paper, we propose a novel navigation algorithm for outdoor environments, which permits robots to travel from one static node to another along a planned path. It utilizes Normal probability weight distribution (NPWD) to assign weights between two nodes dynamically. Heuristics based shortest path (HSP) algorithm is employed to solve complex optimization problems concerned with real-world scenarios. The experiments performed on categorized road databases show significant improvement in timings and complexity of system. Our results justify the effectiveness for the implementation of driver-assist system.

Global positioning system (GPS) and Inertial navigation systems (INS) plays an important role in ... more Global positioning system (GPS) and Inertial navigation systems (INS) plays an important role in vehicle navigation. GPS has limitaions as it sufferes from GPS outagaes sometimes which leads to weaking and loss of signal information. To address these problems, data fusion is carried out using filters which provides better accuracy and predictability of data for navigation purposes. In this paper, GPS/INS datasets were considred. The pre-processed dataset was processed using ∞ filter. This filters bridge the gap of GPS outages, enhancing the accuracy of estimation/prediction of GPS positions. The simulation results show that the accurate GPS positions have been predicted taking into account vehicle measurement data & the trajectory of the vehicle. The entire simulation was carried out using MATLAB

Abstract—Diabetes0can be0mentioned as one of the most lethal and constant si... more Abstract—Diabetes0can be0mentioned as one of the most lethal and constant sicknesses0that may0cause a arise in the glucose levels. Design and development of performance efficient diagnosis tool is very important and plays a vigorous role in initial prediction of a disease and help medical experts and authorities to start with suitable treatment or medication. The insulin produced by pancreases in the subject's body will be affected leading to abnormal behavior and reduction in metabolism of carbohydrates and increases sugar level in the blood and urine. This could impact and damage various body organs such as kidney, heart eyes and nervous system with their normal functionalities. Hence, preliminary stage detection with proper care & medication could reduce the risk of these problems. In the area of medicine to discover patient's data as well as to attain a predictive model or a set of rules, Classification techniques have been continuously used. This study helped diagnose diabetes by selecting three important AI techniques namely the optimal decision tree algorithm model, Type-2 fuzzy expert system and Adaptive Neuro fuzzy inference system which is modified (M-ANFIS). In present research work, these three methods are proposed in order to improve the classification prediction and accuracy. The Pima Indian Diabetes Dataset (PIDD) from UCI Machine Learning Repository dataset was used to carry out validation and predication of the model accuracy. The PIMA dataset was trained and validated for various classification models using MATLAB. The simulation results proved that M-ANFIS based classification provides better accuracy as compared to other methods. Based on the accuracy and validation of simulated results it proved that the proposed model would be helpful as a detection decision tool for Type 2 diabetes prediction and diagnosis.

2020 IEEE International Conference on Electronics, Computing and Communication Technologies (CONECCT), 2020

Coordinate Rotation Digital Computer (CORDIC) algorithm is widely used to improve the efficiency ... more Coordinate Rotation Digital Computer (CORDIC) algorithm is widely used to improve the efficiency of the hardware implementation of the Digital Signal Processing (DSP) algorithms and other mathematical operations. CORDIC based digital signal processing has become an important tool in communications, biomedical, and industrial products. The fundamental downside of the Conventional CORDIC algorithm is that there exists a considerable amount of repetitive iterations which will add up overall delay to the circuit design. In this paper, we propose a CORDIC algorithm, which is based on the reconfigurable CORDIC architectures that can be configured to operate for circular or hyperbolic trajectories in rotation or vectoring-modes and also can perform various trigonometric function (TF) and exponential functions. The calculation is performed either by using rotation mode or vectoring mode. We use Carry Select Adder (CSLA) for extensive reduction of area complication nature over the standard s...

CRC Press eBooks, Feb 9, 2024

2022 International Conference on Industry 4.0 Technology (I4Tech)

2022 IEEE 4th International Conference on Cybernetics, Cognition and Machine Learning Applications (ICCCMLA)

IAES International Journal of Artificial Intelligence (IJ-AI)

Glaucoma is a disease that affects the optic nerve. This disease, over a period of time, can lead... more Glaucoma is a disease that affects the optic nerve. This disease, over a period of time, can lead to loss of vision. Which is known as ‘silent thief of sight’. There are several methods in which the disease can be treated, if detected at an early stage It is not possible for any technology, including artificial intelligence, to replace a doctor. However, it is possible to develop a model based on several classical image processing algorithms, combined with artificial intelligence that can detect onset of glaucoma based on certain parameters of the retinal fundus. This model would play an important role in early detection of the disease and assist the doctor. The traditional methods to detect glaucoma, as efficient as they may be, are usually expensive, a machine learning approach to diagnose from fundus images and accurately classify its severity can be considered to be efficient. Here we propose support vector machine (SVM) method to segregate, train the models using a high-end gra...

2022 IEEE 4th International Conference on Cybernetics, Cognition and Machine Learning Applications (ICCCMLA)

Journal of Engineering

Solid State Technology, Feb 22, 2021

Emerging Research in Computing, Information, Communication and Applications, 2019

Lecture Notes in Electrical Engineering, 2016

Procedia Computer Science, 2016

2020 IEEE International Conference on Electronics, Computing and Communication Technologies (CONECCT), 2020