ABSTRACT
Blurry scenarios, such as light reflections and water ripples, often affect the clarity and signal-to-noise ratio of fish images, posing significant challenges for traditional deep learning models in accurately recognizing fish species. Firstly, deep learning models rely on a large amount of labeled data. However, it is often difficult to label data in blurry scenarios. Secondly, existing deep learning models need to be more effective for the processing of bad, blurry, and otherwise inadequate images, which is an essential reason for their low recognition rate. A method based on the diffusion model and attention mechanism for fish image recognition in blurry scenarios, DiffusionFR, is proposed to solve these problems and improve the performance of species recognition of fish images in blurry scenarios. This paper presents the selection and application of this correcting technique. In the method, DiffusionFR, a two-stage diffusion network model, TSD, is designed to deblur bad, blurry, and otherwise inadequate fish scene pictures to restore clarity, and a learnable attention module, LAM, is intended to improve the accuracy of fish recognition. In addition, a new dataset of fish images in blurry scenarios, BlurryFish, was constructed and used to validate the effectiveness of DiffusionFR, combining bad, blurry, and otherwise inadequate images from the publicly available dataset Fish4Knowledge. The experimental results demonstrate that DiffusionFR achieves outstanding performance on various datasets. On the original dataset, DiffusionFR achieved the highest training accuracy of 97.55%, as well as a Top-1 accuracy test score of 92.02% and a Top-5 accuracy test score of 95.17%. Furthermore, on nine datasets with light reflection noise, the mean values of training accuracy reached a peak at 96.50%, while the mean values of the Top-1 accuracy test and Top-5 accuracy test were at their highest at 90.96% and 94.12%, respectively. Similarly, on three datasets with water ripple noise, the mean values of training accuracy reached a peak at 95.00%, while the mean values of the Top-1 accuracy test and Top-5 accuracy test were at their highest at 89.54% and 92.73%, respectively. These results demonstrate that the method showcases superior accuracy and enhanced robustness in handling original datasets and datasets with light reflection and water ripple noise.
ABSTRACT
With rapid industrial development and population growth, the pollution of soil and groundwater has become a critical concern all over the world. Yet, remediation of contaminated soil and water remains a major challenge. In recent years, apatite has gained a surging interest in environmental remediation because of its high treatment efficiency, low cost, and environmental benignity. This review summarizes recent advances in: (1) natural apatite of phosphate ores and biological source; (2) synthesis of engineered apatite particles (including stabilized or surface-modified apatite nanoparticles); (3) treatment effectiveness of apatite towards various environmental pollutants in soil and groundwater, including heavy metals (e.g., Pb, Zn, Cu, Cd, and Ni), inorganic anions (e.g., As oxyanions and F-), radionuclides (e.g., thorium (Th), strontium (Sr), and uranium (U)), and organic pollutants (e.g., antibiotics, dyes, and pesticides); and (4) the removal and/or interaction mechanisms of apatite towards the different contaminants. Lastly, the knowledge or technology gaps are identified and future research needs are proposed.
ABSTRACT
Batch and column leaching tests were used to study thallium's release and migration behaviour and evaluate its potential toxicity risks in soil. The results indicated that leaching concentrations of Tl using TCLP and SWLP were much higher than the threshold, indicating a high risk of thallium pollution in the soil. Furthermore, the intermittent leaching rate of Tl by Ca2+ and HCl reached its maximum value, demonstrating the easy release of Tl. After HCl leaching, the form of Tl in the soil has changed, and ammonium sulfate has increased its extractability. Additionally, the extensive application of calcium promoted the release of Tl, increasing its potential ecological risk. Spectral analysis showed that Tl was mainly present in minerals such as Kaolinite and Jarosite, and exhibited significant adsorption capacity for Tl. HCl and Ca2+ damaged the crystal structure of the soil, greatly enhancing the migration and mobility of Tl in the environment. More importantly, XPS analysis confirmed that the release of Tl (I) in the soil was the leading cause of increased mobility and bioavailability. Therefore, the results revealed the risk of Tl release in the soil, providing theoretical guidance for its pollution prevention and control.
ABSTRACT
Irradiation at 239 ± 20 nm of a p-H(2) matrix containing methoxysulfinyl chloride, CH(3)OS(O)Cl, at 3.2 K with filtered light from a medium-pressure mercury lamp produced infrared (IR) absorption lines at 3028.4 (attributable to ν(1), CH(2) antisymmetric stretching), 2999.5 (ν(2), CH(3) antisymmetric stretching), 2950.4 (ν(3), CH(3) symmetric stretching), 1465.2 (ν(4), CH(2) scissoring), 1452.0 (ν(5), CH(3) deformation), 1417.8 (ν(6), CH(3) umbrella), 1165.2 (ν(7), CH(3) wagging), 1152.1 (ν(8), S=O stretching mixed with CH(3) rocking), 1147.8 (ν(9), S=O stretching mixed with CH(3) wagging), 989.7 (ν(10), C-O stretching), and 714.5 cm(-1) (ν(11), S-O stretching) modes of syn-CH(3)OSO. When CD(3)OS(O)Cl in a p-H(2) matrix was used, lines at 2275.9 (ν(1)), 2251.9 (ν(2)), 2083.3 (ν(3)), 1070.3 (ν(4)), 1056.0 (ν(5)), 1085.5 (ν(6)), 1159.7 (ν(7)), 920.1 (ν(8)), 889.0 (ν(9)), 976.9 (ν(10)), and 688.9 (ν(11)) cm(-1) appeared and are assigned to syn-CD(3)OSO; the mode numbers correspond to those used for syn-CH(3)OSO. The assignments are based on the photolytic behavior and a comparison of observed vibrational wavenumbers, infrared intensities, and deuterium isotopic shifts with those predicted with the B3P86∕aug-cc-pVTZ method. Our results extend the previously reported four transient IR absorption bands of gaseous syn-CH(3)OSO near 2991, 2956, 1152, and 994 cm(-1) to 11 lines, including those associated with C-O, O-S, and S=O stretching modes. Vibrational wavenumbers of syn-CD(3)OSO are new. These results demonstrate the advantage of a diminished cage effect of solid p-H(2) such that the Cl atom, produced via UV photodissociation of CH(3)OS(O)Cl in situ, might escape from the original cage to yield isolated CH(3)OSO radicals.
