The atmospheric radiation transfer model (RTM) is the foundation and core of the physical retrieval of atmospheric and surface parameters in remote sensing as well as the assimilation of satellite observation data. Infrared RTM is widely used in the retrieval of Earth’s surface temperature, cloud detection, and water vapor remote sensing. This letter is committed to exploring the fast and accurate simulation of atmospheric radiation transfer over the clear-sky ocean using deep learning algorithms, with the key issue being the fast calculation of atmospheric transmittance. We have constructed a neural infrared transmittance model (NITM) for atmospheric radiation transfer simulation from thermal infrared channels and applied it to the visible infrared imaging radiometer suite (VIIRS) M15 and M16 channels. In addition, to improve the model’s performance, we conducted a sensitivity analysis on the inputs and selected predictors with high sensitivity to transmittance. The comparison results with the line-by-line radiative transfer model (LBLRTM) indicate that the algorithm achieves fast and precise simulation of atmospheric radiation transfer, with the simulated brightness temperature (BT) accuracy better than 0.1 K.

This study offers an in-depth exploration into sea surface skin temperature (SSTskin) from the Haiyang-1D (HY-1D) Chinese Ocean Color and Temperature Scanner (COCTS). The main components include intercalibration, cloud detection, and SSTskin retrieval. First, we conduct the intercalibration of COCTS infrared channels utilizing the Visible Infrared Imaging Radiometer Suite (VIIRS) as the reference instrument. A double-differencing methodology is employed to evaluate and correct the COCTS calibration. Next, we introduce a physically based deep learning algorithm for cloud detection, designed to interpret complex textures in satellite imagery. The algorithm demonstrates superior performance across diverse conditions and geographical areas, especially in reducing false flagging of ocean fronts. Finally, we propose an optimal estimation (OE) methodology for COCTS SSTskin retrieval. One focus is on estimating appropriate covariance matrices within the OE algorithm, including an innovative method for dynamically setting the prior SSTskin uncertainty appropriate to local spatial variability. The second focus is to employ the atmospheric correction smoothing algorithm of OE. Both these measures combine to suppress noise and enhance the sensitivity of SSTskin. We assign quality levels to the retrieved SSTskin data. The high-quality COCTS SSTskin is validated using iQuam in situ data. Our results indicate the bias of$- 0.20~^{\circ }$C and the robust standard deviation (RSD) of$0.27~^{\circ }$C between COCTS and in situ SST, with an average sensitivity of 0.87. These findings affirm that the successful implementation of these methodologies significantly enhances the accuracy and reliability of SSTskin data from HY-1D COCTS. This advancement provides substantial benefits to expand the global high-precision SSTskin dataset.

With the evolution of stereoscopic video technology, live video streaming stands at the threshold of a more engaging and immersive future. Apple's recent innovations in video and Extended Reality (XR) technologies have paved the way for immersive live video experiences. This work characterizes Apple's spatial video and explores its potential in live video streaming, providing insights into the future of video streaming applications.

Monocular 6D pose estimation for objects is an essential but challenging task that is commonly applied in computer vision and robotics. Existing two-stage methods solve for rotations with Perspective-n-Point (PnP), which still incorporates translations, resulting in accuracy degeneration. In contrast, direct regression methods adopt Convolutional Neural Networks (CNNs) to solve for rotations and translations jointly but suffer from performance gaps in rotation accuracy. In this article, we propose a novel Normal Vector guided Regression Network (NVR-Net) to directly regress the 6D pose from a single RGB image under the guidance of 3D normal vectors. Specifically, we design a novel Orientation-Aware Feature (OAF) for pose estimation. It consists of two corresponding sets of 3D normal vectors to thoroughly disentangle rotation from translation estimation. Then, we introduce a CNN to predict a dense pixelwise representation of the OAF without viewpoint ambiguity. To estimate rotations and translations individually from the OAF, we propose a novel Pose from Normal Vectors (PNV) head networks under the instruction of a differentiable closed-form solution. Finally, extensive experiments on three common benchmarks demonstrate that our approach outperforms state-of-the-art methods on rotation accuracy and removes the gap between indirect and end-to-end methods. Moreover, our method can estimate the 6D pose of a single object within an RGB image in real-time.

The Haiyang-1C (HY-1C) satellite is the first operational ocean color satellite of the Chinese HY-1 series satellites. The Chinese Ocean Color and Temperature Scanner (COCTS) onboard the HY-1C satellite has 10 channels for ocean color and sea surface temperature (SST) observations. Cloud detection is one of the key preprocessing steps of SST retrieval. In this paper, we use deep neural network U -Net for detection of clouds in HY-1C COCTS imagery. The ground truth of dataset using to train the U -Net model is constructed by Bayesian cloud detection method and manual mask. The overall accuracy has achieved 0.96 on the COCTS test dataset.

The Chinese Ocean Color and Temperature Scanner (COCTS) on board HY-1 series satellites has two thermal infrared channels with the spectrum range of 10.30-11.40 μm and 11.40-12.50 μm for sea surface temperature (SST) observations. To reprocess the Haiyang-1B (HY-1B) COCTS SST, the Bayesian cloud detection and optimal estimation (OE) SST retrieval were applied to COCTS data in this study. The Bayesian cloud detection algorithm that has been developed is based on the Bayes’ theorem and uses simulation of COCTS observations. The MODerate resolution atmospheric TRANsmission (MODTRAN) model was used for simulation of COCTS brightness temperatures. SSTs were retrieved from COCTS by OE from 2009 to 2011 in the northwest Pacific. Comparison of COCTS OE SST with in situ SST showed that the COCTS SSTs are cooler than buoy measurements by –0.23 °C on average, and the standard deviation (SD) of differences was 0.51 °C. A large component of the mean difference is attributable to the cool skin effect at the ocean surface (typically –0.15 to –0.2 °C), the remainder being attributable to simulation and calibration biases. The mean difference of COCTS OE SST with matched skin temperatures from the Advanced Along Track Scanning Radiometer (AATSR) is closer to zero, being –0.09 °C, with a SD of 0.49 °C. These validation results of COCTS OE SST demonstrate that Bayesian cloud detection and OE SST retrieval algorithm work well for improving COCTS SST accuracy, and show the potential of these methods to help develop SST products for operational HY-1 satellites, HY-1C and HY-1D.

Arctic sea surface temperature is an important parameter for the research of climate change monitoring. Satellite remote sensing provides an important means of sea surface temperature (SST) observation, especially when there are few buoy and ship data in the Arctic. The Visible and Infrared Scanning Radiometer (VIRR) onboard the Fengyun-3C (FY-3C) can be used to observe SST. In this paper, intercalibration between VIRR and MODIS is performed to obtain corrected brightness temperature (BT) via double difference method, then Bayesian cloud detection and optimal estimation algorithms based on VIRR corrected BT are used to obtain Arctic VIRR SST. In this paper, only the data in March 2016 are processed, the bias of matchups between VIRR SST and AVHRR SST is -0.24°C, the standard deviation is 0.57°C, the robust standard deviation is 0.45°C. In the future, the data of long time series will then be processed for a complete assessment.

The scanning microwave radiometer (RM) onboard the Haiyang-2A (HY-2A) satellite has low-frequency channels with the capability of observing sea surface temperature (SST) from space. To improve the accuracy of the HY-2A RM SST data, the intercalibration of RM brightness temperature (BT) and SST retrieval were carried out. Based on the simulated BTs using the microwave radiative transfer model (RTM), the double-difference approach was applied to perform the intercalibration of RM BTs for the 10.7, 18.7, 23.8, and 37.0 GHz channels, with the Global Precipitation Measurement (GPM) Microwave Imager (GMI) as the reference. The RM 6.6-GHz BTs were corrected with the RTM modeled BT on account of a lack of the similar channel in GMI. The comparison of RM original BTs with GMI and modeled BTs showed large biases and the strong dependence on latitude. We obtained latitude-dependent coefficients using robust linear regression for RM BT correction and applied recalibrated RM BTs for the SST retrieval. The validation of the RM retrieved SST showed the bias of -0.12 °C and the robust standard deviation (RSD) of 1.10 °C compared with buoy SST in the region between 70°S and 70°N. In the tropical and subtropical regions, the bias was -0.12 °C and the RSD was 0.93 °C. In addition, the relationships between the SST difference and the sea surface and atmospheric parameters were investigated. Both statistics of validation results and error analysis indicated significant improvement of RM SST accuracy.

Haiyang-2 (HY-2) is the first marine dynamic environmental satellite of China, which was launched on August 16, 2011. The scanning microwave radiometer (RM) onboard HY-2 has low-frequency channels with the capability of observing sea surface temperature (SST) from space. In this paper, the Level 2A (L2A) SST products of HY-2 RM are evaluated. The global HY-2 RM L2A SST products are compared with the buoy SST measurements, WindSat SST, and National Oceanic and Atmospheric Administration Optimum Interpolation (OI) weekly SST products for the period from January 2012 to December 2014. The collocations of HY-2 RM, WindSat, and buoy SST data are generated with the spatial window of 0.25° and the temporal window of 0.5 h. The biases are -0.45 °C (RM minus buoy) and -0.41 °C (RM minus WindSat) and the corresponding standard deviations are 1.73 °C and 1.72 °C. The comparisons of the weekly averaged HY-2 RM and OI SST show that the biases of each week difference are from -1.06 °C to 0.48 °C with the mean value of -0.30 °C. The standard deviations of the SST difference are from 0.83 °C to 1.47 °C with the mean value of 1.05 °C. The relationships between SST difference and the sea surface and atmospheric parameters, such as wind speed, wind direction, SST, and water vapor are investigated.

The HY-2 (Haiyang-2) satellite was launched in August 2011. The scanning microwave radiometer (RM) onboard HY-2 has low frequency channels with the capability for sea surface temperature (SST) observations from space. The HY-2 RM L2A SST products are evaluated. The L2A products are compared with WindSat, AMSR2 SST products and in situ SST measurements by buoy. The results indicate a standard deviation around 2oC with fluctuated bias. The comparisons of the daily ascending passes between HY-2 RM and WindSat in January 2013 show the bias of each day is between 0.12oC and 0.75oC and the standard deviation is between 1.75oC and 2.97oC. For descending passes, the bias is from -0.51oC to 0.25oC and standard deviation from 1.56oC to 2.58oC. The statistics of the matchups between HY-2 RM and buoy SST data during the period of August 2012 to December 2014 shows a bias of -0.29oC and standard deviation of 1.94oC. The relationships between SST difference and the surface parameters such as wind speed and SST are investigated.