arXiv subjects

Mandeep Singh

Publications and source records attributed to Mandeep Singh.

12 recordsLinked to original sources

arXiv2026

MinNav: Minimalist Navigation Using Optical Flow For Active Tiny Aerial Robots

Navigation using a monocular camera is pivotal for autonomous operation on tiny aerial robots due to their perfect balance of versatility, cost and accuracy. In this paper, we introduce MinNav, a navigation stack based on optical flow and its uncertainty to fly through a scene with static and dynamic obstacles and unknown-shaped gaps without any prior knowledge of the scene components and/or their locations/ordering. We further improve success rate by using the activeness of the robot to move around in an exploratory way to find obstacles and navigate. We successfully evaluate and demonstrate the proposed approach in many real-world experiments in various environments with static and dynamic obstacles and unknown-shaped gaps with an overall success rate of 70%. To the best of our knowledge, this is the first solution to tackle all the aforementioned navigation cases without prior knowledge using a monocular camera. Our approach is on par in performance with depth based methods with factors of magnitude less computation required and can readily run onboard tiny aerial robots. The accompanying video, supplementary material, code and dataset can be found at https://pear.wpi.edu/research/minnav.html

cs.RO↗

arXiv2026

WheatAI v1.0: An AI-Powered High Throughput Wheat Phenotyping Platform

High-throughput, low-cost phenotyping remains a critical bottleneck in wheat breeding, genetics, and crop management. This is particularly evident in the measurement of complex yield components (i.e., spike and spikelet counts), disease and grain-quality traits related to Fusarium Head Blight (FHB) and Fusarium-Damaged Kernels (FDK), and microscale physiological traits such as density and size of stomata and aperture. We introduce WheatAI (wheatai.net), an AI-powered web application designed to bridge the gap between advanced computer vision, AI and deep learning models, and high-throughput phenotyping (HTP) and practical agricultural applications. WheatAI v1.0 provides an accessible, browser-based interface that supports multiscale data ingestion from smartphones, Unmanned Aerial Vehicles (UAVs), and portable microscopes. The core functionalities of the platform include plot- and field-scale assessment via UAV- and smartphone-based wheat spike detection and counting, as well as smartphone-based spikelet counting. Additionally, it offers grain quality assessment through FDK ratio estimation and kernel morphometric measurements, such as length, width, and area, derived from smartphone images of kernel samples. For leaf-level analysis, WheatAI provides microscale phenotyping through automated stomatal counting, size, and aperture measurement from digital microscopy images. The system supports both single-image and bulk processing via a guided upload-and-run workflow. This platform is designed to reduce labor costs and rater subjectivity while accelerating field-to-lab decision cycles. By providing standardized, image-based outputs, WheatAI enables breeders, agronomists, and producers to implement high-throughput selection and precision scouting at scale.

cs.OH↗

arXiv2025

On the Lieb--Wehrl Entropy conjecture for $SU(N,1)$

We investigate the sharp functional inequalities for the coherent state transforms of $SU(N,1)$. These inequalities are rooted in Wehrl's definition of semiclassical entropy and his conjecture about its minimum value. Lieb resolved this conjecture in 1978, posing a similar question for Bloch coherent states of $SU(2)$. The $SU(2)$ conjecture was settled by Lieb and Solovej in 2014, and the conjecture was extended for a wide class of Lie groups. The generalized Lieb conjecture has been resolved for several Lie groups, including $SU(N),\, N\geq 2$, $SU(1,1)$, and its $AX+B$ subgroup. Under the Li--Su assumption on isoperimetric regions in the complex hyperbolic ball, our sharp functional inequalities for the coherent state transforms extend this resolution to $SU(N,1),\, N\geq 2$. Additionally, we explore the Faber--Krahn inequality, which applies to the short-time Fourier transform with a Gaussian window. This inequality was previously proven by Nicola and Tilli and later extended by Ramos and Tilli to the wavelet transform. In this paper, we further extend this result within the framework of the Bergman space $\mathcal{A}_α$.

math-ph↗

arXiv2025

On the conjecture of non-inner automorphisms of finite $p$-groups with a non-trivial abelian direct factor

Let $p$ be a prime number. A longstanding conjecture asserts that every finite non-abelian $p$-group has a non-inner automorphism of order $p$. In this paper, we prove that the conjecture is true when a finite non-abelian $p$-group $G$ has a non-trivial abelian direct factor. Moreover, we prove that the non-inner automorphism is central and fixes $Φ(G)$ elementwise. As a consequence, we prove that every group which is not purely non-abelian has a non-inner central automorphism of order $p$ which fixes $Φ(G)$ elementwise.

math.GR↗

arXiv2024

On the conjecture of non-inner automorphisms of finite $p$-groups

Let $p$ be a prime number. A longstanding conjecture asserts that every finite non-abelian $p$-group has a non-inner automorphism of order $p$. In this paper, we prove that if $G$ is an odd order finite non-abelian monolithic $p$-group such that every maximal subgroup of $G$ is non-abelian and $[Z(M), g] \leq Z(G)$ for every maximal subgroup $M$ of $G$ and $g \in G \setminus M$. Then $G$ has a non-inner automorphism of order $p$ leaving the Frattini subgroup $Φ(G)$ elementwise fixed.

math.GR↗

arXiv2024

Weighted Fourier inequalities and application of restriction theorems on rank one Riemannian symmetric spaces of noncompact type

This article explores weighted $(L^p, L^q)$ inequalities for the Fourier transform in rank one Riemannian symmetric spaces of noncompact type. We establish both necessary and sufficient conditions for these inequalities to hold. To prove the weighted Fourier inequalities, we apply restriction theorems on symmetric spaces and utilize Calder{ó}n's estimate for sublinear operators. While establishing the necessary conditions, we demonstrate that Harish-Chandra's elementary spherical functions play a crucial role in this setting. Furthermore, we apply our findings to derive Fourier inequalities with polynomial and exponential weights.

math.CA↗

arXiv2020

Emotion Recognition in Audio and Video Using Deep Neural Networks

Humans are able to comprehend information from multiple domains for e.g. speech, text and visual. With advancement of deep learning technology there has been significant improvement of speech recognition. Recognizing emotion from speech is important aspect and with deep learning technology emotion recognition has improved in accuracy and latency. There are still many challenges to improve accuracy. In this work, we attempt to explore different neural networks to improve accuracy of emotion recognition. With different architectures explored, we find (CNN+RNN) + 3DCNN multi-model architecture which processes audio spectrograms and corresponding video frames giving emotion prediction accuracy of 54.0% among 4 emotions and 71.75% among 3 emotions using IEMOCAP[2] dataset.

eess.AS↗

arXiv2019

Regions of existence for a class of nonlinear diffusion type problems

The regions of existence are established for a class of two point nonlinear diffusion type boundary value problems (NDBVP) \begin{eqnarray*} &&\label{abst-intr-1} -s''(x)-ns'(x)-\frac{m}{x}s'(x)=f(x,s), \qquad m>0,~n\in \mathbb{R},\qquad x\in(0,1),\\ &&\label{abst-intr-2} s'(0)=0, \qquad a_{1}s(1)+a_{2}s'(1)=C, \end{eqnarray*} where $a_{1}>0,$ $a_{2}\geq0,~ C\in\mathbb{R}$. These problems arise very frequently in many branches of engineering, applied mathematics, astronomy, biological system and modern science (see \cite{Gatica1989, GRAY1980, Baxley1991, Chandershekhar1939, Duggan1986, Chambre1952}). By using the concept of upper and lower solutions with monotone constructive technique, we derive some sufficient conditions for existence in the regions where $\frac{\partial f}{\partial s}\geq0$ and $\frac{\partial f}{\partial s}\leq0$. Theoretical methods are applied for a set of problems which arise in real life.

math.CA↗

arXiv2017

Nonlinear three point Singular BVPs : A Classification

We analyze the existence of unique solutions of the following class of nonlinear three point singular boundary value problems (SBVPs), \begin{eqnarray*}\label{NL-Singular-P} &&-(x^α y'(x))'= x^αf(x,y),\quad 0 0$, $0<η<1$ and $α\geq 1$. This study shows some novel observations regarding the nature of the solution of the nonlinear three point SBVPs. We observe that when $sup\left(\partial f/\partial y\right)>0$ for $α\in \cup_{n\in \mathbb{N}}\left(4n-1,4n+1\right)$ or $α\in\{1,5,9,\cdots\}$ reverse ordered case occur. When $sup\left(\partial f/\partial y\right)>0$ for $α\in \cup_{n\in \mathbb{N}}\left(4n-3,4n-1\right)$ or $α\in\{3,7,11,\cdots\}$ and when $sup\left(\partial f/\partial y\right)<0$ for all $α\geq 1$ well order case occur.

math.CA↗

arXiv2016

Existence uniqueness for a class of Nonlinear Discrete Boundary Value Problems

A monotone iterative method is proposed to solve nonlinear discrete boundary value problems with the support of upper and lower solutions. We establish some new existence results. Under some sufficient conditions, we establish maximum principle for linear discrete boundary value problem, which relies on Green's function and its constant sign. We then use it to establish existence of unique solution for the nonlinear discrete boundary value problem $Δ^2 y(t-1)= f(t, y(t)),~t\in[1, T]$, $y(0)=0,~y(T+1)=0$.

math.NA↗

arXiv2016

A generalization of POWERS-STØRMER inequality

Let $A,\;B$ be the positive semidefinite matrices. A matrix version of the famous Powers-Størmer's inequality $$2Tr(A^αB^{1-α})\geq Tr(A+B-|A-B|),\;\;\;0\leqα\leq 1,$$ was proven by Audenaert et. al. We establish a comparison of eigenvalues for the matrices $A^αB^{1-α}$ and $A+B-|A-B|, \; 0 \leq α\leq 1,$ subsuming the Powers-Størmer's inequality. We also prove several related norm inequalities.

math.FA↗

arXiv2014

Multipixel sub-shot noise phase measurement with classical light interferometry

We demonstrate accurate phase measurement from low photon level interference data using a constrained optimization method that takes into account the expected redundancy in the unknown phase function. This approach is shown to have significant noise advantage over traditional methods such as balanced homodyning or phase shifting that treat individual pixels in the interference data as independent of each other. Our interference experiments comparing the optimization method with the traditional phase shifting method show that when the same photon resources are used, the optimization method provides phase recoveries with tighter error bars. In particular, RMS phase error performance of the optimization method for low photon number data (10 photons per pixel) shows $>$ 5X noise gain over the phase shifting method. In our experiments where a laser light source is used for illumination, the results imply phase measurement with accuracy better than the conventional single pixel based shot noise limit (SNL) that assumes independent phases at individual pixels. The constrained optimization approach presented here is independent of the nature of light source and may further enhance the accuracy of phase detection when a nonclassical light source is used.

quant-ph↗