DenseNet201Plus: Cost-effective transfer-learning architecture for rapid leaf disease identification with attention mechanisms.

Plant leaf diseases are a significant concern in agriculture due to their detrimental impact on crop productivity and food security. Effective disease management depends on the early and accurate detection and diagnosis of these conditions, facilitating timely intervention and mitigation strategies. In this study, we address the pressing need for accurate and efficient methods for detecting leaf diseases by introducing a new architecture called DenseNet201Plus. DenseNet201 was modified by including superior data augmentation and pre-processing techniques, an attention-based transition mechanism, multiple attention modules, and dense blocks. These modifications enhance the robustness and accuracy of the proposed DenseNet201Plus model in diagnosing diseases related to plant leaves. The proposed architecture was trained using two distinct datasets: Banana Leaf Disease and Black Gram Leaf Disease. Through extensive experimentation, we evaluated the performance of DenseNet201Plus in terms of various classification metrics and achieved values of 0.9012, 0.9012, 0.9012, and 0.9716 for accuracy, precision, recall, and AUC for the banana leaf disease dataset, respectively. Similarly, the black gram leaf disease dataset model provides values of 0.9950, 0.9950, 0.9950, and 1.0 for accuracy, precision, recall, and AUC. Compared to other well-known pre-trained convolutional neural network (CNN) architectures, our proposed model demonstrates superior performance in both utilized datasets. Last but not least, we combined the strength of Grad-CAM++ with our proposed model to enhance the interpretability and localization of disease areas, providing valuable insights for agricultural practitioners and researchers to make informed decisions and optimize disease management strategies.

A robust and light-weight transfer learning-based architecture for accurate detection of leaf diseases across multiple plants using less amount of images.

Leaf diseases are a global threat to crop production and food preservation. Detecting these diseases is crucial for effective management. We introduce LeafDoc-Net, a robust, lightweight transfer-learning architecture for accurately detecting leaf diseases across multiple plant species, even with limited image data. Our approach concatenates two pre-trained image classification deep learning-based models, DenseNet121 and MobileNetV2. We enhance DenseNet121 with an attention-based transition mechanism and global average pooling layers, while MobileNetV2 benefits from adding an attention module and global average pooling layers. We deepen the architecture with extra-dense layers featuring swish activation and batch normalization layers, resulting in a more robust and accurate model for diagnosing leaf-related plant diseases. LeafDoc-Net is evaluated on two distinct datasets, focused on cassava and wheat leaf diseases, demonstrating superior performance compared to existing models in accuracy, precision, recall, and AUC metrics. To gain deeper insights into the model's performance, we utilize Grad-CAM++.

An ab initio Study on the Mechanical Stability, Spin-Dependent Electronic Properties, Molecular Orbital Predictions, and Optical Features of Antiperovskite A3InN (A = Co, Ni).

Structural, mechanical, spin-dependent electronic, magnetic, and optical properties of antiperovskite nitrides A3InN (A = Co, Ni) along with molecular orbital diagram are investigated here by using an ab initio density functional theory (DFT). The mechanical stability, deformation, damage tolerance and ductile nature of A3InN are confirmed from elastic calculations. Different mechanical anisotropy factors are also discussed in detail. The spin dependent electronic properties such as the band structure and density of states (DOS) of A3InN are studied and, the dispersion curves and DOS at Fermi level are different for up and down spins only in case of Co3InN. These calculations also suggest that Co3InN and Ni3InN behave as ferromagnetic and nonmagnetic, respectively. The induced total magnetic moment of Co3InN is found 2.735 µB/cell in our calculation. Mulliken bond population analysis shows that the atomic bonds of A3InN are contributed by both ionic and covalent bonds. Molecular orbital diagrams of A3InN antiperovskites are proposed by analyzing orbital projected band structures. The formation of a molecular orbital energy diagram for Co3InN is similar to Ni3InN with respect to hybridization and orbital sequencing. However, the orbital positions with respect to the Fermi level (E F) and separations between them are different. The Fermi surface of A3InN is composed of multiple nonspherical electron and hole type sheets in which Co3InN displays a spin-dependent Fermi surface. The various ground-state optical functions such as real and imaginary parts of the dielectric constant, optical conductivity, reflectivity, refractive index, absorption coefficient, and loss function of A3InN are studied with implications. The reflectivity spectra reveal that A3InN reflects >45% of incident electromagnetic radiations in both the visible and ultraviolet region, which is an ideal feature of coating material for avoiding solar heating.

RbSnX3 (X = Cl, Br, I): promising lead-free metal halide perovskites for photovoltaics and optoelectronics.

Lead (Pb) free metal halide perovskites by atomistic design are of strong interest to photovoltaics and optoelectronics industries because of the pressing need to resolve Pb-related toxicity and instability challenges. In this study, structural, mechanical, electronic, and optical properties of Pb-free RbSnX3 (X = Cl, Br, I) perovskites have been evaluated by using ab initio density functional theory (DFT) calculations. The computed elastic constants suggest that the Rb-based halide perovskites are mechanically stable and highly ductile, making them suitable as flexible thin films in optoelectronic devices. Besides, the investigated electronic band structures reveal that the RbSnX3 compounds are direct bandgap semiconductors, suitable for photovoltaic and optoelectronic applications. Furthermore, several optical parameters such as dielectric functions, reflectivity, photon absorptions, refractive index, optical conductivity, and loss functions have been investigated and the results predict the excellent optoelectronic efficiency of RbSnX3. Also, the computed mechanical and optical properties of RbSnX3 (X = Cl, Br, I) have been compared with the previously studied CsBX3 (B = Ge, Sn, Pb; X = Cl, Br, I) phases, revealing that the Rb-based perovskites are extremely ductile and possess excellent light absorption and optical conductivity compared to the Cs-based perovskites. Importantly, RbSnI3 shows superior ductility, absorption coefficient, and optical conductivity compared to the CsBX3 (B = Ge, Sn, Pb; X = Cl, Br, I) perovskites. Superior absorption at the ultraviolet region of RbSnI3 holds great promise of this perovskite to be used in next-generation ultraviolet photodetectors.