Dataset Information

Paddynet: An organized dataset of paddy leaves for a smart fertilizer recommendation system.

ABSTRACT: The dataset of Leaf Color Chart (PaddyNet) is publicly unavailable. As far as the author's knowledge, this is the first dataset about paddy leaves based on LCC. This dataset has been generated by collecting images from a particular location such as Sajiali, Dogachia and Shyamnagar at Jashore, Bangladesh. This dataset contains 4 categories of Aman paddy leaves. The leaf images were captured by smart phones. There are 560 images of Aman paddy leaves. The data collection procedure was carried out according to the guidelines of Bangladesh Agricultural Research Institute (BARI). We meticulously categorized the entire dataset with regard to the LCC level and validated the data with the assistance of domain specialists. Hence, the images are analyzed and categorized with standards. The dataset is utilized for recognizing Leaf Color Chart level which will help of farmers recommending nitrogen fertilizer in their paddy fields.

SUBMITTER: Siddique MM

PROVIDER: S-EPMC10477451 | biostudies-literature | 2023 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Paddynet: An organized dataset of paddy leaves for a smart fertilizer recommendation system.

Siddique Md Moradul MM Islam Torikul T Tusher Yeasir Arefin YA Ema Romana Rahman RR Adnan Md Nasim MN Galib Syed Md SM

Data in brief 20230823

The dataset of Leaf Color Chart (PaddyNet) is publicly unavailable. As far as the author's knowledge, this is the first dataset about paddy leaves based on LCC. This dataset has been generated by collecting images from a particular location such as Sajiali, Dogachia and Shyamnagar at Jashore, Bangladesh. This dataset contains 4 categories of Aman paddy leaves. The leaf images were captured by smart phones. There are 560 images of Aman paddy leaves. The data collection procedure was carried out a ...[more]

PMID: 37674508

Similar Datasets

Project description:Conventional farming systems are highly reliant on chemical fertilizers (CFs), which adversely affect soil quality, crop production and the environment. One of the major current challenges of current agriculture is finding ways to increase soil health and crop yield sustainably. Manure application as a substitute for CF is an alternative fertilization strategy for maintaining soil health and biodiversity. However, little is known about the complex response of soil bacterial communities and soil nutrients to manure and CFs application. This study reports the response of soil nutrients, rice yield, and soil microbial community structure to 2 years of continuous manure and CFs application. The study consisted of six treatments: no N fertilizer control (Neg-Con); 100% CF (Pos-Con); 60% cattle manure (CM) + 40% CF (High-CM); 30% CM + 70% CF (Low-CM); 60% poultry manure (PM) + 40% CF (High-PM), and 30% PM + 70% CF (Low-PM). We used high-throughput sequencing of 16S ribosomal RNA gene amplicons to characterize the soil bacterial communities. Results revealed that the addition of manure significantly altered the soil bacterial community composition and structure; and enhanced the relative abundance of phyla Proteobacteria, Chloroflexi, Firmicutes, Acidobacteria, and Planctomycetes. Organic fertilizer treatments, particularly high CM and PM had the highest measured soil bacterial diversity of all treatments. Similarly, integrated application of manure and CFs increased the soil biochemical traits [i.e., pH, total N (TN), soil organic C (SOC), microbial biomass N (MBN), and microbial biomass C (MBC)] and rice grain yield. Average increases in SOC, TN, MBN, and MBC were 43.66, 31.57, 24.34, and 49.45%, respectively, over the years in the High-PM compared with Pos-Con. Redundancy analysis showed that the dominant bacteria phyla were correlated with soil pH, SOC, TN, and microbial biomass, but the relative abundance of Proteobacteria was strongly correlated with environmental factors such as soil pH, SOC, TN, and MBC. We employed a structural equation model to examine the relationship between microbial biomass, soil nutrients and grain yield among treatments. This analysis supported the hypothesis that soil nutrient content and availability directly affect rice grain yield while soil bacteria indirectly affect grain yield through microbial biomass production and nutrient levels. Overall, the findings of this research suggest that the integrated application of CF and manure is a better approach for improving soil health and rice yield.

Dataset Information

Paddynet: An organized dataset of paddy leaves for a smart fertilizer recommendation system.

Publications

Paddynet: An organized dataset of paddy leaves for a smart fertilizer recommendation system.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets