Reflectance spectral dataset for ship coating condition classification using portable Vis/NIR spectroscopy

This dataset contains multispectral reflectance measurements of ship coating surfaces collected using a portable Visible/Near-Infrared (Vis/NIR) spectroscopy system equipped with an AS7265X multispectral sensor. The sensor captures spectral responses across 18 discrete wavelength bands ranging from 410 nm to 940 nm, covering the ultraviolet (UV), visible (VIS), and near-infrared (NIR) spectral regions.

The dataset was developed to support the analysis and classification of ship coating conditions using non-destructive sensing techniques combined with machine learning approaches. Each sample represents a coating inspection point and includes a coating condition label based on the Condition Assessment Program (CAP) classification, along with spectral reflectance intensity values recorded from the AS7265X sensor channels.

The dataset includes four coating condition categories:
- Very Good Condition
- Good Condition
- Class Condition
- Poor Condition

These categories represent different levels of coating integrity and corrosion protection performance. Spectral data were collected under controlled measurement conditions to minimize ambient light interference. Each measurement was repeated multiple times. The resulting spectral dataset enables the investigation of spectral patterns associated with coating degradation, corrosion risk, and variations in surface condition, providing a foundation for developing machine learning models for automated ship coating inspection and condition monitoring.

Data collection was conducted at PT Dok dan Perkapalan Kodja Bahari, specifically at the Galangan Jakarta II shipyard facility in 2024, where coating inspection measurements were conducted on the surfaces of operational vessels.

This research was funded by the Directorate of Research, Technology, and Community Service under the Ministry of Education, Culture, Research, and Technology of the Republic of Indonesia, through Fundamental Research Grant Number: 105/E5/PG.02.00.PL/2024.

本数据集包含舰船涂层表面的多光谱反射率测量数据，采集自搭载AS7265X多光谱传感器的便携式可见/近红外（Visible/Near-Infrared, Vis/NIR）光谱系统。该传感器可捕获18个离散波段的光谱响应信号，波段范围覆盖410 nm至940 nm，涵盖紫外（ultraviolet, UV）、可见（Visible, VIS）及近红外（Near-Infrared, NIR）光谱区域。

本数据集旨在支撑结合机器学习方法的无损检测技术在舰船涂层状态分析与分类任务中的应用。每个样本对应一处涂层检测点位，包含基于涂层评估计划（Condition Assessment Program, CAP）分类体系的涂层状态标签，以及AS7265X传感器各通道记录的光谱反射强度值。

本数据集涵盖四类涂层状态分类：
- 极佳状态
- 良好状态
- 一般状态
- 较差状态

上述分类对应涂层完整性与防腐蚀防护性能的不同等级。光谱数据采集于受控测量环境中，以最大限度降低环境光干扰；每项测量均进行了多次重复。本光谱数据集可用于研究与涂层老化、腐蚀风险及表面状态变化相关的光谱特征规律，为开发用于舰船涂层自动化检测与状态监测的机器学习模型提供了支撑基础。

数据采集工作于2024年在PT Dok dan Perkapalan Kodja Bahari公司的加拉南雅加达二号（Galangan Jakarta II）船厂开展，检测对象为在役舰船的表面涂层。

本研究由印度尼西亚共和国教育、文化、研究与技术部下的研究、技术与社区服务总局通过编号为105/E5/PG.02.00.PL/2024的基础研究资助项目资助。