Small object detection algorithm incorporating swin transformer for tea buds.

Accurate identification of small tea buds is a key technology for tea harvesting robots, which directly affects tea quality and yield. However, due to the complexity of the tea plantation environment and the diversity of tea buds, accurate identification remains an enormous challenge. Current methods based on traditional image processing and machine learning fail to effectively extract subtle features and morphology of small tea buds, resulting in low accuracy and robustness. To achieve accurate identification, this paper proposes a small object detection algorithm called STF-YOLO (Small Target Detection with Swin Transformer and Focused YOLO), which integrates the Swin Transformer module and the YOLOv8 network to improve the detection ability of small objects. The Swin Transformer module extracts visual features based on a self-attention mechanism, which captures global and local context information of small objects to enhance feature representation. The YOLOv8 network is an object detector based on deep convolutional neural networks, offering high speed and precision. Based on the YOLOv8 network, modules including Focus and Depthwise Convolution are introduced to reduce computation and parameters, increase receptive field and feature channels, and improve feature fusion and transmission. Additionally, the Wise Intersection over Union loss is utilized to optimize the network. Experiments conducted on a self-created dataset of tea buds demonstrate that the STF-YOLO model achieves outstanding results, with an accuracy of 91.5% and a mean Average Precision of 89.4%. These results are significantly better than other detectors. Results show that, compared to mainstream algorithms (YOLOv8, YOLOv7, YOLOv5, and YOLOx), the model improves accuracy and F1 score by 5-20.22 percentage points and 0.03-0.13, respectively, proving its effectiveness in enhancing small object detection performance. This research provides technical means for the accurate identification of small tea buds in complex environments and offers insights into small object detection. Future research can further optimize model structures and parameters for more scenarios and tasks, as well as explore data augmentation and model fusion methods to improve generalization ability and robustness.

Uptake epithelia behave in a cell-centric and not systems homeostatic manner in response to zinc depletion and supplementation.

Much remains to be understood about systemic regulation of zinc uptake in vertebrates, and adequate zinc status is far from always achieved in animals or human. In addition to absorbing zinc from the diet, fish are able to take up zinc directly from the water with the gills. This provides an elegant system to study zinc uptake, how it relates to zinc status, and the expression of genes for proteins involved in zinc acquisition. A 21-day experiment was conducted in which zebrafish were acclimated to deficient, control or excess zinc concentrations in the water and feed. Deficient provision of zinc reduced whole body zinc, potassium, sodium and calcium levels whilst zinc concentrations in the uptake epithelia (gills and gut) remained unchanged. Excess levels of zinc caused accumulation of zinc in the gills, intestine and carcass, but impaired whole body iron, sodium and calcium concentrations. Fish subjected to zinc deficiency had, surprisingly, a reduced zinc influx across the gill epithelium, even when tested at a high concentration of zinc in the water. Zinc influx in the excess group was indistinct from the control. Expression of genes for metallothionein-2 (Mt2) and zinc transporters-1, -2, and -8 (Znt1, Znt2, Znt8) in uptake epithelia showed in general a direct relationship with zinc supply, while mRNA for Zip4 was inversely related to zinc supply. Transcripts for the epithelial calcium channel (Ecac/Trpv6) showed time-dependent increased expression in the gills of the deficiency group, and a transient decrease of expression during zinc excess. Transcriptome profiling by microarrays showed that in both gills and intestine, the most markedly affected biological functions were those related to cell growth, proliferation and cancer, closely followed by processes of gene transcription and protein synthesis in general. Whilst changes in zinc supply had profound effects in the intestine on genes associated with uptake and metabolism of macronutrients, many of the unique categories of genes preferentially regulated in the gill could be mapped onto signalling pathways. This included pathways for PPAR/RXR, LXR/RXR, ATM, chemokine, and BMP signalling. Overall, the responses of epithelial tissue to zinc deficiency and excess are best explained by local epithelial homeostasis with no evidence of systemic control.

Dynamic transcriptomic profiles of zebrafish gills in response to zinc supplementation.

BACKGROUND: Dietary zinc supplementation may help to promote growth, boost the immune system, protect against diabetes, and aid recovery from diarrhoea. We exploited the zebrafish (Danio rerio) gill as a unique vertebrate ion transporting epithelium model to study the time-dependent regulatory networks of gene-expression leading to homeostatic control during zinc supplementation. This organ forms a conduit for zinc uptake whilst exhibiting conservation of zinc trafficking components. RESULTS: Fish were maintained with either zinc supplemented water (4.0 µM) and diet (2023 mg zinc kg-1) or water and diet containing Zn2+ at 0.25 µM and 233 mg zinc kg-1, respectively. Gill tissues were harvested at five time points (8 hours to 14 days) and transcriptome changes analysed in quintuplicate using a 16 K microarray with results anchored to gill Zn2+ influx and whole body nutrient composition (protein, carbohydrate, lipid, elements). The number of regulated genes increased up to day 7 but declined as the fish acclimated. In total 525 genes were regulated (having a fold-change more than 1.8 fold change and an adjusted P-value less than 0.1 which is controlling a 10% False discovery rate, FDR) by zinc supplementation, but little overlap was observed between genes regulated at successive time-points. Many genes displayed cyclic expression, typical for homeostatic control mechanisms. Annotation enrichment analysis revealed strong overrepresentation of "transcription factors", with specific association evident with "steroid hormone receptors". A suite of genes linked to "development" were also statistically overrepresented. More specifically, early regulation of genes was linked to a few key transcription factors (e.g. Mtf1, Jun, Stat1, Ppara, Gata3) and was followed by hedgehog and bone morphogenic protein signalling. CONCLUSIONS: The results suggest that zinc supplementation reactivated developmental pathways in the gill and stimulated stem cell differentiation, a response likely reflecting gill remodelling in response to its altered environment. This provides insight to the role of zinc during cell differentiation and illustrates the critical nature of maintaining zinc status. The study also highlights the importance of temporal transcriptomics analysis in order resolve the discrete elements of biological processes, such as zinc acclimation.