.png){kind=icon}
返回
Soil texture analysis using controlled image processing
.png){kind=icon}
OA
AI
DOI：10.1016/j.atech.2024.100588.png)
.png){kind=icon}
.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Soil texture analysis is crucial for crop selection, fertilizer recommendation, and production. Traditional soil testing in the lab using chemicals is highly time-consuming, expensive, and risky harmful chemicals; proper equipment and trained professionals are required to get the readings and to conduct the analysis. These issues can be resolved using image processing. In this study, we proposed a Blackbox prototype machine to take images in a controlled environment under the fixed intensity of light, distance, and standard dry conditions to analyze soil texture. This innovative machine, with its efficient and precise image processing capabilities, has the potential to revolutionize soil texture analysis. Also, we marked the center points of each type of soil texture as defined in the USDA texture triangle. Hundreds of soil samples were prepared for each type according to the center point's sand, silt, and clay ratio. The image processing-based model is trained for texture analysis. This research aims to reduce the soil texture analysis time and provide a system that can do extensive analyses automatically and with accuracy. The proposed Blackbox prototype machine has proven effective in providing a controlled environment for taking images. Also, the proposed model detects soil texture with a maximum accuracy of 99.5%. A proposed model trained on the soil samples of different texture classes available in the USDA texture triangle accurately performed texture analysis. The results benefit the recommendation of appropriate crops and fertilizers based on a given soil sample in a very short time and cost-effectively.
Keyword:
Image processing
Soil texture
Blackbox Prototype
YOLOv8
USDA Texture Triangle
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
5.7
论文数:
2.6K
被引数:
2.5K
机构
引用论文
Soil Surface Texture Classification Using RGB Images Acquired Under Uncontrolled Field Conditions.png)
IEEE ACCESS
OAAI
IEEE ACCESSIF3.6
OAAI分享收藏PVswin-YOLOv8s: UAV-Based Pedestrian and Vehicle Detection for Traffic Management in Smart Cities Using Improved YOLOv8PVswin-YOLOv8s: 基于改进YOLOv8的无人机行人和车辆检测智能城市交通管理DRONESOAAI
DRONESIF4.8
OAAI分享收藏
查看更多内容