2 results
Persistent eutrophication and hypoxia in the coastal ocean
- Minhan Dai, Yangyang Zhao, Fei Chai, Mingru Chen, Nengwang Chen, Yimin Chen, Danyang Cheng, Jianping Gan, Dabo Guan, Yuanyuan Hong, Jialu Huang, Yanting Lee, Kenneth Mei Yee Leung, Phaik Eem Lim, Senjie Lin, Xin Lin, Xin Liu, Zhiqiang Liu, Ya-Wei Luo, Feifei Meng, Chalermrat Sangmanee, Yuan Shen, Khanittha Uthaipan, Wan Izatul Asma Wan Talaat, Xianhui Sean Wan, Cong Wang, Dazhi Wang, Guizhi Wang, Shanlin Wang, Yanmin Wang, Yuntao Wang, Zhe Wang, Zhixuan Wang, Yanping Xu, Jin-Yu Terence Yang, Yan Yang, Moriaki Yasuhara, Dan Yu, Jianmin Yu, Liuqian Yu, Zengkai Zhang, Zhouling Zhang
-
- Journal:
- Cambridge Prisms: Coastal Futures / Volume 1 / 2023
- Published online by Cambridge University Press:
- 23 February 2023, e19
-
- Article
-
- You have access Access
- Open access
- HTML
- Export citation
-
Coastal eutrophication and hypoxia remain a persistent environmental crisis despite the great efforts to reduce nutrient loading and mitigate associated environmental damages. Symptoms of this crisis have appeared to spread rapidly, reaching developing countries in Asia with emergences in Southern America and Africa. The pace of changes and the underlying drivers remain not so clear. To address the gap, we review the up-to-date status and mechanisms of eutrophication and hypoxia in global coastal oceans, upon which we examine the trajectories of changes over the 40 years or longer in six model coastal systems with varying socio-economic development statuses and different levels and histories of eutrophication. Although these coastal systems share common features of eutrophication, site-specific characteristics are also substantial, depending on the regional environmental setting and level of social-economic development along with policy implementation and management. Nevertheless, ecosystem recovery generally needs greater reduction in pressures compared to that initiated degradation and becomes less feasible to achieve past norms with a longer time anthropogenic pressures on the ecosystems. While the qualitative causality between drivers and consequences is well established, quantitative attribution of these drivers to eutrophication and hypoxia remains difficult especially when we consider the social economic drivers because the changes in coastal ecosystems are subject to multiple influences and the cause–effect relationship is often non-linear. Such relationships are further complicated by climate changes that have been accelerating over the past few decades. The knowledge gaps that limit our quantitative and mechanistic understanding of the human-coastal ocean nexus are identified, which is essential for science-based policy making. Recognizing lessons from past management practices, we advocate for a better, more efficient indexing system of coastal eutrophication and an advanced regional earth system modeling framework with optimal modules of human dimensions to facilitate the development and evaluation of effective policy and restoration actions.
Automatic food detection in egocentric images using artificial intelligence technology
- Wenyan Jia, Yuecheng Li, Ruowei Qu, Thomas Baranowski, Lora E Burke, Hong Zhang, Yicheng Bai, Juliet M Mancino, Guizhi Xu, Zhi-Hong Mao, Mingui Sun
-
- Journal:
- Public Health Nutrition / Volume 22 / Issue 7 / May 2019
- Published online by Cambridge University Press:
- 26 March 2018, pp. 1168-1179
-
- Article
-
- You have access Access
- HTML
- Export citation
-
Objective
To develop an artificial intelligence (AI)-based algorithm which can automatically detect food items from images acquired by an egocentric wearable camera for dietary assessment.
DesignTo study human diet and lifestyle, large sets of egocentric images were acquired using a wearable device, called eButton, from free-living individuals. Three thousand nine hundred images containing real-world activities, which formed eButton data set 1, were manually selected from thirty subjects. eButton data set 2 contained 29 515 images acquired from a research participant in a week-long unrestricted recording. They included both food- and non-food-related real-life activities, such as dining at both home and restaurants, cooking, shopping, gardening, housekeeping chores, taking classes, gym exercise, etc. All images in these data sets were classified as food/non-food images based on their tags generated by a convolutional neural network.
ResultsA cross data-set test was conducted on eButton data set 1. The overall accuracy of food detection was 91·5 and 86·4 %, respectively, when one-half of data set 1 was used for training and the other half for testing. For eButton data set 2, 74·0 % sensitivity and 87·0 % specificity were obtained if both ‘food’ and ‘drink’ were considered as food images. Alternatively, if only ‘food’ items were considered, the sensitivity and specificity reached 85·0 and 85·8 %, respectively.
ConclusionsThe AI technology can automatically detect foods from low-quality, wearable camera-acquired real-world egocentric images with reasonable accuracy, reducing both the burden of data processing and privacy concerns.