Aravind Padmasanan
Research Subjects
Aravind Padmasanan has a total of 2 co-authors. Here are the top co-authors by number of shared publications:
- Praveen Pankajakshan
Quantitative Image Analysis Unit, Institut Pasteur, 75015 Paris, France
2 shared publications - Subbiah Sundar
Centre for Computational Mathematics and Data Science Department of Mathematics IIT Madras 600036 Chennai India
2 shared publications
Latest Publications
- Open Access
Deep Architectures Fail to Generalize: A Lightweight Alternative for Agricultural Domain Transfer in Hyperspectral Images
We present a novel framework for hyperspectral satellite image classification that explicitly balances spatial nearness with spectral similarity. The proposed method is trained on closed-set datasets, and it generalizes well to open-set agricultural scenarios that include both class distribution shifts and presence of novel and absence of known classes. This scenario is reflective of real-world agricultural conditions, where geographic regions, crop types, and seasonal dynamics vary widely and labeled data are scarce and expensive. The input data are projected onto a lower-dimensional spectral manifold, and a pixel-wise classifier generates an initial class probability saliency map. A kernel-based spectral-spatial weighting strategy fuses the spatial-spectral features. The proposed approach improves the classification accuracy by – over spectral-only models on benchmark datasets. Incorporating an additional unsupervised learning refinement step further improves accuracy, surpassing several recent state-of-the-art methods. Requiring only 1– labeled training data and at most two tuneable parameters, the framework operates with minimal computational overhead, qualifying it as a data-efficient and scalable few-shot learning solution. Recent deep architectures although exhibit high accuracy under data rich conditions, often show limited transferability under low-label, open-set agricultural conditions. We demonstrate transferability to new domains—including unseen crop classes (e.g., paddy), seasons, and regions (e.g., Piedmont, Italy)—without re-training. Rice paddy fields play a pivotal role in global food security but are also a significant contributor to greenhouse gas emissions, especially methane, and extent mapping is very critical. This work presents a novel perspective on hyperspectral classification and open-set adaptation, suited for sustainable agriculture with limited labels and low-resource domain generalization.Open AccessDeep Architectures Fail to Generalize: A Lightweight Alternative for Agricultural Domain Transfer in Hyperspectral Images
We present a novel framework for hyperspectral satellite image classification that explicitly balances spatial nearness with spectral similarity. The proposed method is trained on closed-set datasets, and it generalizes well to open-set agricultural scenarios that include both class distribution shifts and presence of novel and absence of known classes. This scenario is reflective of real-world agricultural conditions, where geographic regions, crop types, and seasonal dynamics vary widely and labeled data are scarce and expensive. The input data are projected onto a lower-dimensional spectral manifold, and a pixel-wise classifier generates an initial class probability saliency map. A kernel-based spectral-spatial weighting strategy fuses the spatial-spectral features. The proposed approach improves the classification accuracy by – over spectral-only models on benchmark datasets. Incorporating an additional unsupervised learning refinement step further improves accuracy, surpassing several recent state-of-the-art methods. Requiring only 1– labeled training data and at most two tuneable parameters, the framework operates with minimal computational overhead, qualifying it as a data-efficient and scalable few-shot learning solution. Recent deep architectures although exhibit high accuracy under data rich conditions, often show limited transferability under low-label, open-set agricultural conditions. We demonstrate transferability to new domains—including unseen crop classes (e.g., paddy), seasons, and regions (e.g., Piedmont, Italy)—without re-training. Rice paddy fields play a pivotal role in global food security but are also a significant contributor to greenhouse gas emissions, especially methane, and extent mapping is very critical. This work presents a novel perspective on hyperspectral classification and open-set adaptation, suited for sustainable agriculture with limited labels and low-resource domain generalization. - Open Access
- 26 November 2025
- Published by MDPI AG in Preprints.org
Deep Architectures Fail to Generalize: A Lightweight Alternative for Agricultural Domain Transfer in Hyperspectral Images
We present a novel framework for hyperspectral satellite image classification that explicitly balances spatial nearness with spectral similarity. The proposed method is trained on closed-set datasets, and it was tested on zero-shot agricultural scenarios that include both class distribution shifts, and presence of novel and absence of known classes. This scenario is reflective of real-world agricultural conditions, where geographic regions, crop types, and seasonal dynamics vary widely and labeled data are scarce and expensive. The input data are projected onto a lower-dimensional spectral manifold, and a pre-trained pixel-wise classifier generates an initial class probability saliency map. A kernel-based spectral-spatial weighting strategy fuses the spatial-spectral features. The proposed approach improves the classification accuracy by 7.22%–15% over spectral-only models on benchmark datasets after iterative convergence. Incorporating the additional unsupervised learning and weak labeling helped surpass several recent state-of-the-art methods. Requiring only 1%–10% labeled training data and at most two tuneable parameters, the framework operates with minimal computational overhead, qualifying it as a data-efficient and scalable learning solution. Recent deep architectures, although exhibiting high closed-set accuracies, often show limited transferability under low-label, open-set or zero-shot agricultural conditions where class distributions shift and novel classes emerge. The rice crop play a pivotal role in global food security but are also a significant contributor to greenhouse gas emissions, especially methane, and extent mapping is very critical. We demonstrate transferability to new domains-including unseen crop class (e.g. , paddy), seasons, and regions (e.g. , Piedmont, Italy)—without re-training. This work presents a novel perspective on hyperspectral classification and domain transferability, suited for sustainable agriculture with limited labels and low-resource domain generalization.
- 26 November 2025
- Published by MDPI AG in Preprints.org
Open AccessDeep Architectures Fail to Generalize: A Lightweight Alternative for Agricultural Domain Transfer in Hyperspectral Images
We present a novel framework for hyperspectral satellite image classification that explicitly balances spatial nearness with spectral similarity. The proposed method is trained on closed-set datasets, and it was tested on zero-shot agricultural scenarios that include both class distribution shifts, and presence of novel and absence of known classes. This scenario is reflective of real-world agricultural conditions, where geographic regions, crop types, and seasonal dynamics vary widely and labeled data are scarce and expensive. The input data are projected onto a lower-dimensional spectral manifold, and a pre-trained pixel-wise classifier generates an initial class probability saliency map. A kernel-based spectral-spatial weighting strategy fuses the spatial-spectral features. The proposed approach improves the classification accuracy by 7.22%–15% over spectral-only models on benchmark datasets after iterative convergence. Incorporating the additional unsupervised learning and weak labeling helped surpass several recent state-of-the-art methods. Requiring only 1%–10% labeled training data and at most two tuneable parameters, the framework operates with minimal computational overhead, qualifying it as a data-efficient and scalable learning solution. Recent deep architectures, although exhibiting high closed-set accuracies, often show limited transferability under low-label, open-set or zero-shot agricultural conditions where class distributions shift and novel classes emerge. The rice crop play a pivotal role in global food security but are also a significant contributor to greenhouse gas emissions, especially methane, and extent mapping is very critical. We demonstrate transferability to new domains-including unseen crop class (e.g. , paddy), seasons, and regions (e.g. , Piedmont, Italy)—without re-training. This work presents a novel perspective on hyperspectral classification and domain transferability, suited for sustainable agriculture with limited labels and low-resource domain generalization.