Vase-Life Monitoring System for Cut Flowers Using Deep Learning and Multiple Cameras DOI Creative Commons

Ji Yeong Ham,

Yong-Tae Kim, Suong Tuyet Thi Ha

и другие.

Plants, Год журнала: 2025, Номер 14(7), С. 1076 - 1076

Опубликована: Апрель 1, 2025

Here, we developed a vase-life monitoring system (VMS) to automatically and accurately assess the post-harvest quality vase life (VL) of cut roses. The VMS integrates camera imaging with YOLOv8 (You Only Look Once version 8) deep learning algorithm continuously monitor major physiological parameters including flower opening, fresh weight, water uptake, gray mold disease incidence. Our results showed that can measure main factors roses by obtaining precise consistent data. values measured for physiology closely correlated those observation (OBS). Additionally, achieved high performance in model an object detection accuracy 90%. mAP0.5 supported evaluating VL Regression analysis revealed strong correlation between VL, VMS, OBS. incorporating microscope detected early stages development. These show plant is highly effective method using could also be applied breeding process, which requires rapid measurements important characteristics species, such as resistance, develop superior cultivars.

Язык: Английский

Impacts of extreme climate events on vegetation succession at the northern foothills of Yinshan mountain, inner Mongolia DOI Creative Commons
Pingping Zhou,

Zilong Liao,

Xiaoyan Song

и другие.

Frontiers in Environmental Science, Год журнала: 2025, Номер 13

Опубликована: Март 11, 2025

Extreme climate events significantly impact vegetation ecosystems in dry regions, particularly areas adjacent to the northern foothills of Yinshan Mountain (NYSM). However, there remains limited understanding how responds such events. Analyzing response regions drought is beneficial for protection and restoration ecosystem. This study analyzes spatiotemporal variation characteristics extreme NDVI. By employing correlation analysis geographic detectors, it explores NDVI The findings indicate a recent decline temperature concurrent rise precipitation From 2000 2020, demonstrated consistent improvement, trend expected persist future. exhibited strong negative with NDVI, whereas positive correlation. Furthermore, possess greater explanatory power variability compared research provide theoretical basis different types NYSM respond events, they inform targeted ecological measures based on varying responses these

Язык: Английский

Процитировано

0
Detecting Tasseling Rate of Breeding Maize Using UAV-based RGB Images and STB-YOLO Model DOI Creative Commons

Boyi Tang,

Jingping Zhou, Xiaolan Li

и другие.

Smart Agricultural Technology, Год журнала: 2025, Номер unknown, С. 100893 - 100893

Опубликована: Март 1, 2025

Язык: Английский

Процитировано

0
Enhancing yield prediction in maize breeding using UAV-derived RGB imagery: a novel classification-integrated regression approach DOI Creative Commons
Haixiao Ge, Qi Zhang, Min Shen

и другие.

Frontiers in Plant Science, Год журнала: 2025, Номер 16

Опубликована: Март 20, 2025

Accurate grain yield prediction is crucial for optimizing agricultural practices and ensuring food security. This study introduces a novel classification-integrated regression approach to improve maize using UAV-derived RGB imagery. We compared three classifiers—Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF)—to categorize data into low, medium, high classes. Among these, SVM achieved the highest classification accuracy was selected classifying prior regression. Two methodologies were evaluated: Method 1 (direct RF on full dataset) 2 (SVM followed by class-specific regression). Multi-temporal vegetation indices (VIs) analyzed across key growth stages, with early vegetative phase yielding lowest errors. significantly outperformed 1, reducing RMSE 45.1% in calibration (0.28 t/ha vs. 0.51 t/ha) 3.3% validation (0.89 0.92 t/ha). integrated framework demonstrates advantage of combining precise estimation, providing scalable tool breeding programs. The results highlight potential UAV-based phenotyping enhance productivity support global systems.

Язык: Английский

Процитировано

0
Enhancing winter wheat plant nitrogen content prediction across different regions: Integration of UAV spectral data and transfer learning strategies DOI
Zongpeng Li,

Qian Cheng,

Li Chen

и другие.

Computers and Electronics in Agriculture, Год журнала: 2025, Номер 234, С. 110322 - 110322

Опубликована: Март 23, 2025

Язык: Английский

Процитировано

0
Vase-Life Monitoring System for Cut Flowers Using Deep Learning and Multiple Cameras DOI Creative Commons

Ji Yeong Ham,

Yong-Tae Kim, Suong Tuyet Thi Ha

и другие.

Plants, Год журнала: 2025, Номер 14(7), С. 1076 - 1076

Опубликована: Апрель 1, 2025

Here, we developed a vase-life monitoring system (VMS) to automatically and accurately assess the post-harvest quality vase life (VL) of cut roses. The VMS integrates camera imaging with YOLOv8 (You Only Look Once version 8) deep learning algorithm continuously monitor major physiological parameters including flower opening, fresh weight, water uptake, gray mold disease incidence. Our results showed that can measure main factors roses by obtaining precise consistent data. values measured for physiology closely correlated those observation (OBS). Additionally, achieved high performance in model an object detection accuracy 90%. mAP0.5 supported evaluating VL Regression analysis revealed strong correlation between VL, VMS, OBS. incorporating microscope detected early stages development. These show plant is highly effective method using could also be applied breeding process, which requires rapid measurements important characteristics species, such as resistance, develop superior cultivars.

Язык: Английский

Процитировано

0