Germination of chili pepper seeds is critical for crop yield and resource utilization. A high germination rate increases yield and effectively reduces resource wastage. This study collected 450 macroscopic images of chili pepper seeds and constructed a dataset for deep learning training through standardized germination experiments. Six deep learning models were evaluated to improve the chili pepper seed classification accuracy and germination rate. After comparing the performance of the models, MobileNet_v2 performed the best, not only having the fewest number of parameters but also achieving a 98.89% accuracy and 97.82% F1 score. The model improved the original germination rate from 87.33% to 100% on the test set, significantly optimizing the seed selection process

Crop diseases seriously affect food security, and traditional identification methods are inefficient and inaccurate. This paper proposes a GoogLeNet model with an attention mechanism. By integrating an attention module inside the Inception module, the recognition ability of subtle disease features and complex backgrounds is improved. Based on strict data preprocessing and enhancement, the proposed method achieves 87.75% accuracy on the AI Challenger 2018 crop disease dataset, which is better than the existing advanced methods, which verifies the effectiveness and practicability of the method and provides technical support for smart agriculture.

Animal pattern recognition from nighttime grayscale images is crucial for wildlife protection and ecological monitoring. Most of the current models suffer from a large parameter scale, making them unsuitable for deployment in resource-constrained environments. To address this challenge, this study proposes a multi-layer knowledge distillation approach based on the teacher Convnext model to improve the lightweight student model's classification performance effectively. Experimental results show that the parameters of the distilled CifarResNet20 model are only 0.27M, and the accuracy is 88.76%, which is superior to the traditional single-layer distillation and another tiny student model. The study confirms the efficiency and practical value of the proposed method in practical applications such as ecological monitoring.

In this work, iron oxide (Fe3O4) nanoparticles were deposited on multi-walled carbon nanotubes (MWNTs) by a simple chemical coprecipitation method and Fe3O4-decorated MWNTs (Fe-MWNTs)/polypyrrole (PPy) nanocomposites (Fe-MWNTs/PPy) were prepared by oxidation polymerization. The effect of the PPy on the electrochemical properties of the Fe-MWNTs was investigated. The structures characteristics and surface properties of MWNTs, Fe-MWNTs, and Fe-MWNTs/PPy were characterized by X-ray diffraction and X-ray photoelectron spectroscopy, respectively. The electrochemical performances of MWNTs, Fe-MWNTs, and Fe-MWNTs/PPy were determined by cyclic voltammetry and galvanostatic charge/discharge characteristics in a 1.0 M sodium sulfite electrolyte. The results showed that the Fe-MWNTs/PPy electrode had typical pseudo-capacitive behavior and a specific capacitance significantly greater than that of the Fe-MWNT electrode, indicating an enhanced electrochemical performance of the Fe-MWNTs/PPy due to their high electrical properties.

Petroleum pitch-based activated carbons (ACs) were obtained in this work from a combination of pretreatment with different amounts of potassium permanganate (KMnO4) and chemical activation with potassium hydroxide. The surface characteristics of the pitch after the KMnO4 pretreatment were characterized by means of Fourier transform infrared spectroscopy (FT-IR). The structural characteristics of the pitch after the KMnO4 pretreatment were determined by means of X-ray diffraction. The influence of the KMnO4 treatment on the textural properties of the petroleum pitch-based ACs was investigated by means of N2/77K adsorption isotherms. The investigation also involved the use of the Brunauer-Emmett-Teller equation and the Dubinin-Radushkevich method. The FT-IR results show that the pretreatment promotes the formation of surface oxygen functionalities and leads to an increase of the interplanar distance (d002) of the functional groups induced between carbon layers. Moreover, the specific surface area of the pitch-based ACs increases in proportion to the amount of KMnO4 pretreatment and reaches its highest value of 2334 m2/g with 2 g of KMnO4 because the surface oxygen groups of the pitch act as an active site during chemical activation.

In thermal cutting process, gas flow injected from the nozzle has a significant effect on the cutting materials. The gas flow is difficult to observe gases are transparent, therefore, in this study, Schlieren method was adopted to visualize the gas flow inside the kerf. The kerf shape was inserted between two slices of transparent glass in order to imitate the real cutting environment. In order to get the flow characteristics, a high speed camera was equipped and the image processing was applied to compare the before and after injection images. As a result, the method for visualizing gas flow was successfully developed and also expected to be applied to the analysis of purging gas in various welding environments.

Cynanchum wilfordii Radix (CWR), Arctium lappa L (ALL), and Dioscorea opposite (DO) have been known to improve blood lipid profile, blood pressure, and inflammation. To find the optimal combination ratio of CWR, ALL, and DO in terms of vascular health improvement, we compared the effects of various combinations on gene expression of Vascular cell adhesion protein 1 (VCAM-1) in human aortic smooth muscle cells (HASMC). VCAM-1 mediates endothelial leukocyte adhesion and is upregulated in atherosclerosis. Cells was stimulated by TNF-α (10 ng/㎖, 2h) and treated with various combinations for 24 h. A combination (CADM5, CWR:ALL:DO = 2:1:1) showed the strongest suppression of VCAM-1 so that CADM5 was chosen for further experiments. We performed cell viability test with CADM5 (0, 3.125, 12.5, 25, 50, and 100 ㎍/㎖) and no cytotoxicity was found. We also investigated the effect of CADM5 on protein expression of VCAM-1, ICAM-1, Nrf-2, and HO-1 using western blotting. We found that CADM5 diminished the expression of VCAM-1 and increased the expression of Nrf-2 and HO-1. Therefore, we concluded that CADM5 (CWR:ALL:DO = 2:1:1) effectively improves vascular health by regulating the expression of VCAM-1.

According to the Marine Traffic Safety Law, revised in 2009, Marine Traffic Safety Audit is introduced to secure the safe navigation, to prevent the marine accident and to maximize the efficiency of the port. In this audit system, marine traffic safety assessment is the most important scheme because the primary purpose of the audit system is to identify potential risk elements affecting safe navigation. Even though the reliability of audit result depends on the selection of assessment models, there are no independent assessment models for Korean coastal waters and most of models used in Korea currently are developed by foreign countries. Therefore, the development of the independent assessment model for Korean coastal water is required. This study, prior to the development of independent assessment model, aims to provide a basic data by comparing two foreign assessment models in Ulsan port area with marine accident statistics data.