Pest recognition in the agriculture field plays a critical issue for the farmers which diminishes economic growth. So far, traditional practices were followed by farmers to increase yield production. Nowadays, researchers execute a deep learning approach to classify various kinds of images practically. In this paper, Deep convolutional neural networks (DCNN) are used to recognise ten kinds of pests present in the paddy crop. The data repository contains around 3549 pest images that affect the paddy crops, Since Deep Learning supports well for larger data-set so the data augmentation process is carried out. The neural model is build using various kinds of DCNN architecture, interpretation was made over the models based on the accuracy rate and the performance. The transfer learning approach is applied over the pest data set by fine-tuning the hyperparameters and the layers of the ResNet-50 model. By comparing the resultant value, the fine-tuned ResNet-50 model produced better accuracy of 95.012% among other models. The obtained resultant value describes the effective performance of the model in pest disease classification.

In soil texture classification, accuracy with interpretability is the key to sustainable agriculture and environmental management. The presented ATFEM (Advanced Triptych Feature Engineering and Modeling framework) framework synergizes handcrafted texture features with learned deep representations through a three-stream architecture: VGG-RTPNet (Residual Texture-Preserving Network based on Visual Geometry Group-16) for texture, ResNet-DANet (Residual Network integrated with Dual Attention Network) for semantics, and Swin-FANet (Shifted Window-based Frequency-Aware Network based on Transformer) for spectral spatial correlation. Subsequently, these branches help in extracting fine-grained structural, dual-attention-enhanced semantic, and spectral-spatial correlation-wise features of soil-image data. To further eliminate redundancy from the feature sets and arrive at the best representation, a Feature Fusion and Selection strategy employing an enhanced hybrid metaheuristic method termed EWJFO (Enhanced Wombat-Jellyfish Feature Optimization) is proposed. It synthesizes the adaptive exploration behavior of Wombat Optimization Algorithm (WOA) with the swift control convergence tempo of the Jellyfish Search Optimizer (JSO) to select the best feature subset. In addition, a new handcrafted descriptor for soil texture image analysis referred as Farthing Ornament of Histogram of Oriented Gradients (F-HOG) has been introduced with adapative. Conventional HOG is burdened with having high-dimensional redundancy and hence suffers from noise sensitivity, F-HOG combines the effect of a Butterworth frequency filter to remove the unwanted high-frequency artifacts and then goes on to perform the statistical selection of the most frequent gradient bins, thus reducing dimensions and retaining quite a bit of the discriminative structural information. The experiments were conducted on a self-built soil texture image dataset consisting of 4,000 labeled images distributed among five texture classes. ATFEM achieved an accuracy of 98.10%, an F1 score of 89.60%, Cohen’s kappa rating of 94.80%, and an AUC of 98.10%, outperforming cutting-edge methods such as CatBoost-DNN, GBDT-CNN, and SVC-RF. This work offers an upscalable, explainable, and expressively accurate solution for soil texture mapping in precision agriculture and environmental monitoring.

Breast cancer is a tumor caused by the excessive growth of cells in the breast tissue or near the region of the breast. Breast cancer is most commonly found among women, and it starts developing when the cell tissues from lump of the breast become abnormal or when there is a calcium deposit in the breast. These affected cells form a large lump that consequently becomes a tumor. Digital infrared images are obtained based on the metabolism of the breast and vascular circulation of the blood flow in and around the breast region, which has more visibility than the normal breast region. In this article, we diagnose breast cancer by processing a thermal image that is acquired from thermal cameras. By analyzing the information, we can implement image processing steps to predict quantitative and qualitative information. In this work, we propose a hybrid clustering algorithm with distance measurements. The clustering step includes adaptive fuzzy k-means clustering with Chebyshev distance with improved classifiers, which include neural networks. The article discusses the experimental results along with the comparison using various metrics, such as accuracy, time, and error rates.