With the development of industrialization and urbanization, the consumption and pollution of water resources are becoming more and more serious. Water quality monitoring is an extremely important technical means to protect water resources. However, the current popular water quality monitoring methods have their shortcomings, such as a low signal-to-noise ratio of satellites, poor time continuity of unmanned aerial vehicles, and frequent maintenance of in situ underwater probes. A non-contact near-surface system that can continuously monitor water quality fluctuation is urgently needed. This study proposes an automatic near-surface water quality monitoring system, which can complete the physical equipment construction, data collection, and processing of the application scenario, prove the feasibility of the self-developed equipment and methods and obtain high-performance retrieval results of four water quality parameters, namely chemical oxygen demand (COD), turbidity, ammoniacal nitrogen (NH3-N), and dissolved oxygen (DO). For each water quality parameter, fourteen machine learning algorithms were compared and evaluated with five assessment indexes. Because the ensemble learning models combine the prediction results of multiple basic learners, they have higher robustness in the prediction of water quality parameters. The optimal determination coefficients ($R^2$) of COD, turbidity, NH3-N, and DO in the test dataset are 0.92, 0.98, 0.95, and 0.91, respectively. The results show the superiority of near-surface remote sensing, which has potential application value in inland, coastal, and various water bodies in the future.

As an important part of the "air–ground" integrated water quality monitoring system, the inversion of water quality from unmanned airborne hyperspectral image has attracted more and more attention. Meanwhile, unmanned aerial vehicles (UAVs) have the characteristics of small size, flexibility and quick response, and can complete the task of water environment detection in a large area, thus avoiding the difficulty in obtaining satellite data and the limitation of single-point monitoring by ground stations. Most researchers use UAV for water quality monitoring, they take water samples back to library or directly use portable sensors for measurement while flying drones at the same time. Due to the UAV speed and route planning, the actual sampling time and the UAV passing time cannot be guaranteed to be completely synchronized, and there will be a difference of a few minutes. For water quality parameters such as chromaticity (chroma), chlorophyll-a (chl-a), chemical oxygen demand (COD), etc., the changes in a few minutes are small and negligible. However, for the turbidity, especially in flowing water body, this value of it will change within a certain range. This phenomenon will lead to noise error in the measured suspended matter or turbidity, which will affect the performance of regression model and retrieval accuracy. In this study, to solve the quality problem of label data in a flowing water body, an unmanned airborne hyperspectral water quality retrieval experiment was carried out in the Xiao River in Xi’an, China, which verified the rationality and effectiveness of label denoising analysis of different water quality parameters. To identify noisy label instances efficiently, we proposed an instance selection scheme. Furthermore, considering the limitation of the dataset samples and the characteristic of regression task, we build a 1DCNN model combining a self attention mechanism (SAM) and the network achieves the best retrieving performance on turbidity and chroma data. The experiment results show that, for flowing water body, the noisy-label instance selection method can improve retrieval performance slightly on the COD parameter, but improve greatly on turbidity and chroma data.

IRI estimation using analysis of dynamic tire pressure and axle acceleration road roughness detection;International Roughness Index (IRI);dynamic tire pressure;mobile sensing;road health monitoring;Pavement Condition Index (PCI); A new method is developed to estimate road profile in order to estimate IRI based on the ASTM standard. This method utilizes an accelerometer and a Dynamic Tire Pressure Sensor (DTPS) to estimate road roughness. The accelerometer measures the vertical axle acceleration. The DTPS, which is mounted on the tire's valve stem, measures dynamic pressure inside the tire while driving. Calibrated transfer functions are used to estimate road profile using the signals from the two sensors. A field test was conducted on roads with different quality conditions in the city of Brockton, MA. The IRI values estimated with this new method match the actual road conditions measured with Pavement Condition Index (PCI) based on the ASTM standard, images taken from an onboard camera and passengers' perceptions. IRI has negative correlation with PCI in general since they have overlapping features. Compared to the current method of IRI measurement, the advantage of this method is that a) the cost is reduced; b) more space is saved; c) more time is saved; and d) mounting the two sensors are universally compatible to most cars and vans. Therefore, this method has the potential to provide continuous and global monitoring the health of roadways.

Underwater optical images are severely degraded by veiling backscatter, low signal-to-noise ratio, and unstable polarization measurements, especially in turbid active-illumination environments. This paper proposes a single-shot active polarimetric denoising and descattering method for underwater image restoration using a division-of-focal-plane polarization camera. Instead of estimating target-dependent polarization parameters through iterative optimization, the active polarization model is simplified by setting the target-light polarization degree to the endpoint of its valid interval. The restored target signal is then obtained by comparing the observed degree of polarization with a spatially varying backscattering-polarization field. This field is estimated from reliable target-free water regions through robust surface fitting, and can also be inferred from low-reflectance black-pixel samples when explicit water background is unavailable. To suppress noise-induced false polarization information, DoFP-BM3D denoising is introduced before Stokes reconstruction and polarization-parameter estimation. Experiments on low-polarization, high-polarization, and mixed-polarization targets under different turbidity levels demonstrate that the proposed method provides stronger background suppression, clearer target structures, and more stable quantitative performance than classical enhancement methods, conventional polarization descattering models, and recent physics-based approaches. Additional analyses show that red polarized illumination and front-end denoising further improve restoration reliability under severe scattering. The proposed framework offers a physically interpretable, non-iterative, and acquisition-efficient solution for robust underwater polarimetric perception in complex scattering environments.

Underwater vision is degraded by wavelength-dependent absorption and scattering, making conventional intensity imaging unreliable for target detection and material discrimination. We report a compact snapshot spectro-polarimetric imaging system that simultaneously acquires spectral and polarization cues in a single exposure. The system stacks a 4 × 4 multispectral filter array (MSFA) with a division-of-focal-plane (DoFP) polarization sensor, providing 16 spectral channels across 430–550 nm and four linear polarization states within a 29 mm × 29 mm × 29 mm prototype. A guard-pixel extraction strategy is used to suppress spectral boundary crosstalk, and the raw mosaic frame is decoded into wavelength-resolved intensity, Stokes, and polarization-parameter images. Tank experiments in turbid water demonstrate two task-oriented capabilities: spectro-polarimetric fusion improves shell-target extraction by suppressing background leakage, while band-wise polarization descattering followed by spectral-angle-mapper (SAM) classification enhances discrimination of visually similar green materials. These results show that snapshot spectro-polarimetric imaging provides a compact and synchronous route for complex underwater vision tasks.