Robust deep learning method for fruit decay detection and plant identification: enhancing food security and quality control.

This paper presents a robust deep learning method for fruit decay detection and plant identification. By addressing the limitations of previous studies that primarily focused on model accuracy, our approach aims to provide a more comprehensive solution that considers the challenges of robustness and limited data scenarios. The proposed method achieves exceptional accuracy of 99.93%, surpassing established models. In addition to its exceptional accuracy, the proposed method highlights the significance of robustness and adaptability in limited data scenarios. The proposed model exhibits strong performance even under the challenging conditions, such as intense lighting variations and partial image obstructions. Extensive evaluations demonstrate its robust performance, generalization ability, and minimal misclassifications. The inclusion of Class Activation Maps enhances the model's capability to identify distinguishing features between fresh and rotten fruits. This research has significant implications for fruit quality control, economic loss reduction, and applications in agriculture, transportation, and scientific research. The proposed method serves as a valuable resource for fruit and plant-related industries. It offers precise adaptation to specific data, customization of the network architecture, and effective training even with limited data. Overall, this research contributes to fruit quality control, economic loss reduction, and waste minimization.

Enhanced photocatalytic degradation of methylene blue using a novel counter-rotating disc reactor.

Introduction: This research introduces an innovative photocatalytic reactor designed to address challenges in wastewater treatment, with a focus on enhancing dye degradation and reducing Chemical Oxygen Demand (COD). Methods: The reactor is designed with counter-rotational movements of discs to enhance hydrodynamics and mass transfer, along with a 3D-printed, interchangeable component system to boost efficacy. TiO2 nanoparticles, composed of 80% anatase and 20% rutile, are thermally immobilized on glass discs. The effectiveness of various treatment variables was assessed through a Central Composite Design (CCD), guided by a Response Surface Methodology (RSM) model. Results: The RSM analysis reveals that the linear, quadratic, and interactive effects of the counter-rotational movements significantly influence the efficiency of dye and COD removal. The RSM model yields coefficients of determination (R2) values of 0.9758 and 0.9765 for the predictive models of dye and COD removal, respectively. Optimized parameters for dye removal include a pH of 6.05, disc rotation speed of 22.35 rpm, initial dye concentration of 3.15 × 10-5 M, residence time of 7.98 h, and the number of nanoparticle layers set at 3.99, resulting in 96.63% dye removal and 65.81% COD removal under optimal conditions. Discussion: Notably, the reactor demonstrates potential for efficient treatment within a near-neutral pH range, which could reduce costs and resource use by eliminating the need for pH adjustments. The implementation of discs rotating in opposite directions marks a significant advancement in the process of dye removal.

Degradation of direct red 23 dye in a plug flow reactor with adjustable angle baffles using nano-Fe3O4: modelling and optimisation.

Photocatalytic degradation is a key technique in wastewater treatment, particularly for toxic dye removal, yet challenges related to poor hydrodynamics and mass transfer limitations persist. This study addresses these challenges by innovatively employing an adjustable-angle baffle in a plug flow reactor (PFR) to enhance dye removal efficiency. A lab-scale PFR with an adjustable baffle was utilised to assess the impact of various factors, including baffle angle, catalyst concentration, hydraulic retention time (HRT), pH, and initial dye concentration, on the removal of direct red 23 dye. The experimental design employed a central composite design (CCD), with subsequent data analysis using response surface methodology (RSM) and artificial neural network (ANN) models. The findings demonstrate that the adjustable baffle significantly impacts dye removal, achieving maximum efficiency at an optimal angle of 77.5 degrees. The ANN model outperformed the RSM model, with a higher determination coefficient (R2) of 0.994 compared to 0.928. Furthermore, RSM and genetic algorithms yielded closely aligned optimal conditions, validating their accuracy. The optimised conditions achieved a dye removal efficiency of 89.47%. Significantly, the study also identified degradation as the dominant mechanism over adsorption and highlighted the impressive stability of nano-Fe3O4 during the recycling process. Mineralisation analysis revealed the presence of lightweight organic residual molecules post-treatment. These outcomes demonstrate the effectiveness of adjustable baffles in PFRs, marking a significant advancement in wastewater treatment technologies and underscoring the critical role of baffle orientation and catalyst concentration in optimising dye removal processes.