Object detection is a challenging and important field in the domain of computervision. Automated detection and counting of the number of siliques non-destructively, which is the most important yield component trait i...
详细信息
ISBN:
(纸本)9789819752119;9789819752126
Object detection is a challenging and important field in the domain of computervision. Automated detection and counting of the number of siliques non-destructively, which is the most important yield component trait in indian mustard (B. juncea), which is important as well as widely cultivated oilseed in India, is done here. We have used Deep Learning based one-stage object detection model, YOLOv5 in the counting of mustard siliques from the images captured by a DSLR camera (SONY Alpha 7iii, 24.2MP) manually mounted on a tripod. A GUI-based desktop software, MuSiC v1.0 is developed using Tkinter, Python backed by YOLOv5 model developed from annotated training dataset using Roboflow annotator. Approximately 22,800 annotations were used to train the deep learning YOLOv5 model and the whole dataset was partitioned into training, testing and validation in the ratio of 8:1:1. The Confusion matrix, F1 score and mAP showed a moderate result of 66%, 0.51 and 0.38, respectively, with the handcrafted ground truth data. The deployed model in MuSiC v1.0 showed a promising result of approximately good accuracy for silique count. This is the first initiative of creation of a desktop application for mustard silique count non-destructively.
Rice is a popular staple diet in India, and its demand has recently increased. Thanjavur, located in the Cauvery Delta region, is known as the rice granary of South India. Due to recent technological advancements, dig...
详细信息
ISBN:
(纸本)9789819752119;9789819752126
Rice is a popular staple diet in India, and its demand has recently increased. Thanjavur, located in the Cauvery Delta region, is known as the rice granary of South India. Due to recent technological advancements, digital farming and globalization have significantly impacted the agricultural industry. It is crucial to differentiate between types of rice grains to prevent fraudulent labeling during import and export. To achieve this, a dataset, namely "TaPaSe Dataset", comprising five varieties of rice, including MTU 1010, MTU 1290, Narmadha, Pacha Ponni, and Sonna Masur, which are mainly cultivated in Thanjavur, has been collected. We designed an image acquisition system to capture the aforementioned varieties in real time. The captured paddy rice images are highly challenging in the sense that all the images are captured under illumination and scale variations. We evaluated existing deep learning models to understand their ability to classify paddy seed varieties. The existing pre-trained models attain remarkable recognition rates on the proposed paddy seed varieties dataset.
The proceedings contain 99 papers. The topics discussed include: robust watermarking scheme based on multiresolution fractional Fourier transform;an adaptive clustering based non-linear filter for the restoration of i...
ISBN:
(纸本)9780769534763
The proceedings contain 99 papers. The topics discussed include: robust watermarking scheme based on multiresolution fractional Fourier transform;an adaptive clustering based non-linear filter for the restoration of impulse corrupted digital images;visibility cuts: a system for rendering dynamic virtual environments;multi-scale method for adaptive mesh editing based on rigidity estimation;automatic perspective camera calibration based on an incomplete set of chessboard markers;an algebraic framework for discrete basis functions in computervision;regularization of incompletely, irregularly and randomly sampled data;fast and secure real-time video encryption;quantization based data hiding scheme for efficient quality access control of images using DWT via lifting;document image segmentation as a spectral partitioning problem;integrated detect-track framework for multi-view face detection in video;and a framework for analysis of surveillance videos.
The proceedings contain 49 papers. The topics discussed include: enhancing face quality assessment through age and expression analysis;IDA-UIE: an iterative framework for deep network based degradation aware underwate...
ISBN:
(纸本)9798400710759
The proceedings contain 49 papers. The topics discussed include: enhancing face quality assessment through age and expression analysis;IDA-UIE: an iterative framework for deep network based degradation aware underwater image enhancement;an improved framework for precision grading of renal cell carcinoma using histopathological images;lost in context: the influence of context on feature attribution methods for object recognition;vision-language modeling with regularized spatial transformer networks for all weather crosswind landing of aircraft;on the efficacy of guidance tasks in panoptic segmentation;enhancing generalization ability in deepfake detection via continual learning;and a computervision framework on biomechanical analysis of jump landings.
The proceedings contain 80 papers. The topics discussed include: PredGAN - a deep multi-scale video prediction framework for detecting anomalies in videos;multiple kernel fisher discriminant metric learning for person...
ISBN:
(纸本)9781450366151
The proceedings contain 80 papers. The topics discussed include: PredGAN - a deep multi-scale video prediction framework for detecting anomalies in videos;multiple kernel fisher discriminant metric learning for person re-identification;vision-based steering angle prediction by the fusion of depth and intensity deep features;zero-shot learning using graph regularized latent discriminative cross-domain triplets;perfectly secure Shamir’s secret sharing scheme for privacy preserving imageprocessing over cloud;moving average recurrent neural network model for video-based person re-identification;activity recognition in egocentric videos using bag of key action units;and learning end-to-end autonomous driving using guided auxiliary supervision.
The proceedings contain 81 papers. The topics discussed include: are buildings only instances? exploration in architectural style categories;geometry directed browser for personal photographs;heritage app: annotating ...
ISBN:
(纸本)9781450316606
The proceedings contain 81 papers. The topics discussed include: are buildings only instances? exploration in architectural style categories;geometry directed browser for personal photographs;heritage app: annotating images on mobile phones;content level access to digital library of India pages;large-scale statistical modeling of motion patterns: a Bayesian non-parametric approach;salient object detection using a fuzzy theoretic approach;a finite mixture model based on pair-copula construction of multivariate distributions and its application to color image segmentation;local appearance based robust tracking via sparse representation;semi-supervised multiple instance learning based domain adaptation for object detection;a grammar-based GUI for single view reconstruction;accelerating non-local denoising with a patch based dictionary;and viewpoint based mobile robotic exploration aiding object search in indoor environment.
The proceedings contain 70 papers. The topics discussed include: the human action image and its application to motion recognition;human action recognition in video by 'meaningful' poses;human action recognitio...
ISBN:
(纸本)9781450300605
The proceedings contain 70 papers. The topics discussed include: the human action image and its application to motion recognition;human action recognition in video by 'meaningful' poses;human action recognition using a dynamic Bayesian action network with 2D part models;unsupervised discovery of activity correlations using latent topic models;optimal shot detection and recognition using Shiryaev-Roberts statistics;the human action image and its application to motion recognition;a novel technique for sketch to photo synthesis;implicit surface octrees for ray tracing point models;a multi-view extension of the ICP algorithm;a convex multi-view stereo formulation with robustness to occlusions and time-varying clutter;realtime motion segmentation based multibody visual slam;image based PTM synthesis for realistic rendering of low resolution 3D models;and reground-background separation on GPU using order based approaches.
The proceedings contain 49 papers. The topics discussed include: teaching GANs to sketch in vector format;iterative gradient encoding network using feature co-occurrence loss for single image reflection removal;MSDNet...
ISBN:
(纸本)9781450391276
The proceedings contain 49 papers. The topics discussed include: teaching GANs to sketch in vector format;iterative gradient encoding network using feature co-occurrence loss for single image reflection removal;MSDNet: a novel multi-stage progressive image dehazing network;monocular multi-layer layout estimation for warehouse racks;cross-view kernel similarity metric learning using pairwise constraints for person re-identification;abnormality detection and classification of macular diseases from optical coherence tomography images: using feature space comparison;a novel unsupervised thresholding technique for Landsat image change detection;robust brain state decoding using bidirectional long short term memory networks in functional MRI;NTU-X: an enhanced large-scale dataset for improving pose-based recognition of subtle human actions;and interpretive self-supervised pre-training: boosting performance on visual medical data.
The proceedings contain 39 papers. The topics discussed include: single view homography estimation for an inclined textured planar surface: overcoming the inverse and ill-posed challenge!;MMAG: mutually motivated atte...
ISBN:
(纸本)9798400716256
The proceedings contain 39 papers. The topics discussed include: single view homography estimation for an inclined textured planar surface: overcoming the inverse and ill-posed challenge!;MMAG: mutually motivated attention gates for simultaneous extraction of contextual and spatial information from a monocular image;automatic assessment of communication skill in real-world job interviews: a comparative study using deep learning and domain adaptation;an efficient motor imagery classification framework using sparse brain connectivity and class-consistent dictionary learning from electroencephalogram signals;mandala as computational art: vectorization and beyond;degradation aware multi-scale approach to no reference image quality assessment;dual stage semantic information based generative adversarial network for image super-resolution;knowledge distillation with ensemble calibration;and a novel framework for robust fingerprint representations using deep convolution network with attention mechanism.
暂无评论