Author

Hardik Modi

18 papers · 2012–2026

2012peak 3 in 20132026

Frequent co-authors

Papers

showing 1–18 of 18 RSS

Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone

Marah Abdin, Jyoti Aneja, Hany Awadalla +126 more

cs.CL · April 2024 · cited 2,528× · arXiv ↗ · more like this ·

We introduce phi-3-mini, a 3.8 billion parameter language model trained on 3.3 trillion tokens, whose overall performance, as measured by both academic benchmarks and internal testing, rivals that of models such as Mixtral 8x7B and GPT-3.5 (e.g., phi-3-mini achieves 69% on MMLU and 8.38 on MT-bench), despite being small enough to be deployed on a phone. Our training dataset is a scaled-up version of the one used for phi-2, composed of heavily filtered publicly available web data and synthetic data. The model is also further aligned for robustness, safety, and chat format. We also provide parameter-scaling results with a 7B, 14B models trained for 4.8T tokens, called phi-3-small, phi-3-medium, both significantly more capable than phi-3-mini (e.g., respectively 75%, 78% on MMLU, and 8.7, 8.9 on MT-bench). To enhance multilingual, multimodal, and long-context capabilities, we introduce…

5G technology of mobile communication: A survey

A. Gohil, Hardik Modi, Shobhit K. Patel

Computer Networks and Communications · March 2013 · cited 303× · DOI ↗ · more like this ·

The objective of this paper is comprehensive study related to 5G technology of mobile communication. Existing research work in mobile communication is related to 5G technology. In 5G, researches are related to the development of World Wide Wireless Web (WWWW), Dynamic Adhoc Wireless Networks (DAWN) and Real Wireless Communication. The most important technologies for 5G technologies are 802.11 Wireless Local Area Networks (WLAN) and 802.16 Wireless Metropolitan Area Networks (WMAN), Ad-hoc Wireless Personal Area Network (WPAN) and Wireless networks for digital communication. 4G technology will include several standards under a common umbrella, similar to 3G, but with IEEE 802.xx wireless mobile networks integrated from the commencement. The major contribution of this paper is the key provisions of 5G (Fifth Generation) technology of mobile communication, which is seen as consumer…

Leukemia Detection using Digital Image Processing Techniques

P. Himali, Hardik Modi, Manoj Pandya +1 more

Computer Vision and Pattern Recognition · International Journal of Applied Information Systems · 4 Nov 2015 · cited 43× · DOI ↗ · more like this ·

This paper discusses about methods for detection of leukemia. Various image processing techniques are used for identification of red blood cell and immature white cells. Different disease like anemia, leukemia, malaria, deficiency of vitamin B12, etc. can be diagnosed accordingly. Objective is to detect the leukemia affected cells and count it. According to detection of immature blast cells, leukemia can be identified and also define that either it is chronic or acute. To detect immature cells, number of methods are used like histogram equalization, linear contrast stretching, some morphological techniques like area opening, area closing, erosion, dilation. Watershed transform, K means, histogram equalization & linear contrast stretching, and shape based features are accurate 72.2%, 72%, 73.7 % and 97.8% respectively.

Body Posture Correction and Hand Gesture Detection Using Federated Learning and Mediapipe

Rijul Singhal, Hardik Modi, Sahana Srihari +3 more

Human-Computer Interaction · 3 Mar 2023 · cited 7× · DOI ↗ · more like this ·

In today’s post-covid culture, where everyone works from home, there is a huge possibility of serious long-term health problems. A lot of people have started taking up exercises at home and if done incorrectly, they can have major negative effects. Another one of the main contributors to these health issues is bad sitting posture, which is only exacerbated when working for hours on end. Hand gesture detection has many useful applications in elderly healthcare, automating actions and gesture-based presentations and games. To help users with these actions, our paper proposes pinpointing the points of the error to the user in real-time and in a lightweight manner for yoga posture correction. The incorrect positions shall be shown in real-time on top of the user’s video feed to help them correct it properly. The user shall be told about when they are sitting in a bad position, and the…

Comparatives study of Various Techniques against Camcorder Piracy in Theater

Nilesh Dubey, Hardik Modi

Computer Vision and Pattern Recognition · December 2018 · cited 7× · DOI ↗ · more like this ·

For the long decay the cinema industries are suffering from camcorder piracy due to that every year cinema industries are losing billions of dollars. Most important cause of piracy is camcorder piracy, in which pirates record the movie during playback in theater. DCI (Digital cinema Initiative) given many rule and regulation to protect from piracy but still it is increasing, reason is there is no concrete technical solution exist. In this paper various techniques that can be opted in future for fighting against the piracy is taken. There are two types of solution exist for the overcoming the piracy problem one is through deterring the pirate and other is not to let it be captured through camera. Watermarking is one way to deter pirates and watermark can be embedded in frames or audio of videos. Modulations techniques can be used to generate flicker that cannot be detected by HVS but…

A Novel Approach to Detect Chronic Leukemia using Shape based Feature Extraction and Identification with Digital Image Processing

Himali Vaghela, Hardik Modi, Manoj Pandya +1 more

Computer Vision and Pattern Recognition · International Journal of Applied Information Systems · 6 Oct 2016 · cited 7× · DOI ↗ · more like this ·

In this paper, some shape based features like area, perimeter, roundness, standard deviation etc. are used to recognize different types of white blood cells like monocyte, lymphocytes, eosinophil, basophil, neutrophils etc. Using image processing techniques, result can be obtained within 3-4 minute. To perform shape base features operation, contrast of RGB image has to be increased for better detection of white cells. After recognition of each and every cell, classification is performed to detect either it is CML (Chronic Myelogenous Leukemia) or CLL (chronic Lymphocytic leukemia). This algorithm is performed on 30 images. Out of 30, it is successful on 28 images. So it gives accuracy of 93.33%.

A Robust Discrete Wavelet Transform Based Adaptive Watermarking Scheme in YCbCr Color Space against Camcorder Recording in Cinema/Movie Theates

Nilesh Dubey, Hardik Modi

Computer Vision and Pattern Recognition · Engineered Science · 2021 · cited 6× · DOI ↗ · more like this ·

Camcorder-based piracy is increasing nowadays because of the high quality of camcorder devices. The most affected industries are theatre and cinema. To defeat theatre piracy which causes huge loss to the motion picture industry of the world, here in this paper, a robust discrete wavelet transform (DWT) based video watermarking approach has been presented in which a payload formed with date-time of the show, registration number of theatres, and six-digit zonal Improvement code (ZIP) of the theatre location embedded in the movie before it plays back in the theatre. To assure that the video and the algorithm are of good quality and robustness, the adaptive watermarking pattern has been used to minimize the watermarking area in the host video. In this study, YCbCr color space has been adopted to use luminance factor of the image to watermark because the luminance part of the image is the…

Comprehensive Study and Comparative Analysis of Different Types of Background Sub-traction Algorithms

Priyank Shah, Hardik Modi

Computer Vision and Pattern Recognition · International Journal of Image Graphics and Signal Processing · 28 Apr 2014 · cited 6× · DOI ↗ · more like this ·

There are many methods proposed for Background Subtraction algorithm in past years. Background subtraction algorithm is widely used for real time moving object detection in video surveillance system. In this paper we have studied and implemented different types of methods used for segmentation in Background subtraction algorithm with static camera. This paper gives good understanding about procedure to obtain foreground using existing common methods of Background Subtraction, their complexity, utility and also provide basics which will useful to improve performance in the future . First, we have explained the basic steps and procedure used in vision based moving object detection. Then, we have debriefed the common methods of background subtraction like Simple method, statistical methods like Mean and Median filter, Frame Differencing and W4 System method, Running Gaussian Average and…

Comprehensive study of the features, execution steps and microarchitecture of the superscalar processors

Nirali Shah, Y. H. Shah, Hardik Modi

Hardware and Architecture · December 2013 · cited 5× · DOI ↗ · more like this ·

The paper introduces the concept of Superscalar processors; with details about the unique features which help achieve the final goal of lesser execution time and increased speed with minimal of added complexity in the architecture. The paper begins with comparing the other types of processors like Vector and Scalar, and explains how Superscalars prove better ones and further it also excoriates the common misconception about Superpipelined and Superscalar methods of instruction execution. Adding on, the unique features of Superscalars, which are Out of Order execution of instructions, Register Renaming, etc. are explained with example. In addition to features, the basic six implementation steps are introduced, and three basic tasks of Superscalars are explained which include Parallel Decoding, Superscalar Instruction and Parallel Instruction. Further on, the Microarchitecture of…

Comprehensive Analysis of Deep Learning-based Human Activity Recognition approaches based on Accuracy

Aniruddh G. Fataniya, Hardik P. Modi

Computer Vision and Pattern Recognition · International Journal of Computing and Digital Systems · 31 Oct 2022 · cited 4× · DOI ↗ · more like this ·

Human Activity Recognition (HAR) is a vital area of Computer Vision. HAR focuses on various activities carried out by humans. Information relative to the human activities is collected by smart sensors and wearable devices. HAR is classified into two categories, e.g. (a) Vision-based, i.e. human activities are captured in form of image and video and (b) Sensor-based, i.e. human activity input can be taken from wearable devices and object tagging techniques. Human activity recognition is an extensive thrust area for Content-based video analysis, Human-machine interaction, animation, healthcare fields. The paper presents a comprehensive analysis of various deep learning-based approaches adopted to implement human activity recognition based on accuracy. It is observed that for the vision-based category the performance of the Depth Camera-based Recurrent Neural Network model is 99.55%…

Comparative Study of HSV Color Model and Ycbcr Color Model to Detect Nucleus of White Cells

Himali Vaghela, Hardik Modi, Manoj Pandya +1 more

Computer Vision and Pattern Recognition · International Journal of Computer Applications · 15 Sep 2016 · cited 2× · DOI ↗ · more like this ·

Main objective of this paper is to extract nucleus of white cells using image processing techniques.Here, nucleus of white cells are extracted from images using HSV color space and YCbCr color space. Using both of method, comparison between two methods has been checked. It has been proved that YCbCr is better than HSV color model by some experiment. Here, experiment is done on 15 images. HSV color model is given accurate result only on 5 images out of 15 and YCbCr color is given accurate model only on13 images out of 15. So accuracy of HSV and YCbCr model is 33.34% and 86.67% respectively. Here, white cell nucleus detection is useful to detect, blood cancer or Leukemia. It reduce processing time of pathologist and give result in short period of time.

Obstacle Avoidance System in Autonomous Ground Vehicle Using Ground Plane Image Processing and Omnidirectional Drive

Hardik Modi, V. Shastri Nirav, H. Shastri Malav

Computer Vision and Pattern Recognition · February 2015 · cited 2× · DOI ↗ · more like this ·

Today one of the major concerns for any industry is availability of workspace and adequate intratransport systems. This paper suggests the possible and probably cheap solution for both of above incapacities by implying sensor controlled motion of Omni-directional wheels. We have synchronized the image processing capability of Raspberry Pie processor with the sensitivity of motor driver circuit to obtain a precise and accurate motion of platform mounted with mecanum wheel that do not require any turning radius. As an image sensor we have used the simple CCD camera. The set of Image transformations are applied on the image to detect the obstacle precisely. We designed mecanum wheel with main constraint kept as shortage of space and load carrying capacity, hence we were able to report design of the wheel with specific dimensions. The image processing and simulation is done in the MATLAB…

Comprehensive study of non touch enable password system

Monica Varia, Hardik Modi

Human-Computer Interaction · IJARCCE · 27 Dec 2014 · cited 1× · DOI ↗ · more like this ·

Since computer has been invented, people are using many devices to interact with computer and devices such as tablets, smart phones, etc. In the current age of technology, the user is expected to touch or non touch the machine interface to give input. Hand Gesture, where natural hand or red tape is used to communicate without touching machine interface. It gives a feeling of natural way of interacting with the devices. This paper presents the techniques used to interact with machine interface. Different non touch enable technologies through which user can give input to the particular system. Also introduces to the red color detection technology which is cost effective. This technology can be used in real time applications for password system in the areas like ATM, Jewellers Shops, Home Security, etc. This can be an attempt to replace stylus and keyboards with the natural finger without…

Design and implementation of a dynamic power-aware secure processing architecture on FPGA using VERILOG HDL

Hardik P. Modi, Taoufik Saidani, Mayur Makwana +5 more

Artificial Intelligence · Ain Shams Engineering Journal · 2 Sep 2026 · cited 0× · DOI ↗ · more like this ·

This paper describes a power-aware adaptive security architecture for resource-constrained FPGA edge devices. Always on cryptographic modules waste energy during normal operation, so this design activates a 16-bit PRESENT lightweight block cipher only when the input data looks anomalous. A 4-tap symmetric FIR filter smooths the signal in real time, and a dedicated anomaly detector watches the output. Once filtered data crosses an absolute threshold, a BUFGCE integrated clock-gating cell routes the system clock to the secure module and encryption begins. The design was synthesized and implemented on a Xilinx Spartan-6 FPGA at 50 MHz. It’s compact 62 LUTs and 64 flip-flops and the BUFGCE-based clock gating cuts unnecessary switching activity, which should improve dynamic power efficiency. Together, the results show that this event-driven, clock-gated approach can deliver solid…

Implementation Of CORDIC Algorithms For Fundamental Calculation

P. D. Patel, Hardik Modi

Computational Theory and Mathematics · International journal of advance research and innovative ideas in education · 2016 · cited 0× · OpenAlex ↗ · more like this ·

CORDIC or Coordinate Rotation Digital Computer is a faster and easy to implement and powerful algorithm used in real time computations, trigonometric calculation. There are two mode in CORDIC Algorithm one is rotating mode and other is vectoring mode. Here We use CORDIC algorithm for fundamental calculation like triangular calculation using Xlinx 12.1 and FPGA Spartan 3E.CORDIC has use simple shift and add operation there is no multiplication and division is there. Here we use pipelined CORDIC algorithm for better performance. At the last Final results has been compared with base paper. Simple Multiplier is use to calculate simple mathematical calculation for more than two dimension and spectrum calculation generally not possible that’s why here we design CORDIC algorithm.

Investigation and Implementation of Square Type Patch Based Algorithm on Texture Synthesis of Image Inpainting Technique for Medical Images

Hardik Modi, Mandar Sahasrabuddhe

Computer Vision and Pattern Recognition · 31 Jan 2013 · cited 0× · DOI ↗ · more like this ·

In this paper, we have investigated and implemented “Square Type Patch Based Algorithm of Texture Synthesis” for Medical Images. The algorithm is designed to unravel the texture synthesis dilemma of Image Inpainting Technique. It proposes to overlay together patches of an input image in order to construct the output image. The uniqueness of the algorithm is in allowing neighboring patches to meet on boundaries that are not purely straight lines. We will exclusively confer the parameters that organize the performance of the algorithm and the investigation's outcome that we obtained when implementing. We have used Different Types of Human Cells Images as Medical Images and implemented results over them. Human Cells have repetitive structure. So, we have taken Hybrid Stem Cells, Nerve Cells, Stem Cells and Endothelial Cells as Medical Images and generated the texture synthesized results of…

Implementation of Image in Painting Technique using Different Types of PDEs

Hardik Modi, Mandar Sahasrabudhe

Computer Vision and Pattern Recognition · CiiT international journal of digital image processing · 2012 · cited 0× · OpenAlex ↗ · more like this ·

The objective of this paper is to implement image inpainting technique using different types of PDEs. Image Inpainting is a technique of modifying an image in an undetectable form. Its most often used to repair an image, although it can easily be used to remove unwanted objects. The modification of images in a way that is non-detectable for an observer who does not know the original image is a practice as old as artistic creation itself. An effective technique for image inpainting has been developed based on partial differential equation (PDE). Instead of solving the problem in frequency domain, this rather new approach evaluates images in time domain. The basic concept starts from the impression of diffusion as a physical process and draws an analogy between the image inpainting process and the diffusion. Images can be comparable to heat, fluid, and gas which spontaneously move from…

Implementation of 8-Channel Temperature Scanner using Programmable System on Chip 3

Varun A. Patel, Vrushank A. Patel, Meet Patel +2 more

Computer Networks and Communications · CiiT international journal of programmable device circuits and systems · 2012 · cited 0× · OpenAlex ↗ · more like this ·

Today‟s industries depend on automation on most of their operation. Automation requires the measurement of analog signals. This makes analog to digital conversion one of the important fields in automation industries as microcontrollers which are the heart of the automation operates on digital data. In this application we are going to measure temperature from eight different channels which then can be processed to control the operation of the device based on the application. The temperature is converted to analog voltages using the temperature sensors. The processor unit we are using is Cypress PSoC3 (Programmable System on Chip 3) which is an embedded system that has on-chip 8051 microcontroller along with digital and analog peripherals. The PSoC3 has on chip Delta Sigma ADC which we are going to utilize for the conversion of analog voltage to digital voltage. The measured voltage is…