With the development of multi-spectral imaging techniques, many new multi-spectral imaging devices have been developed in recent years. Red-green-blue and near-infrared (RGBN) cameras are widely used because they capture visible light and near-infrared light simultaneously, but they inevitably introduce color desaturation. Because there is clear multicollinearity among the RGBN channels, the ordinary least squares regression (OLSR) color correction method performs poorly. To solve color bias and multicollinearity, an RGBN camera color correction pipeline is proposed. A large number of nonlinear regression color correction methods that consist of combinations of four regression methods and nine nonlinear transforms are evaluated in this study. The results show that the proposed OLSR-based compound transform color correction method and partial least-squares regression (PLSR) based Gaussian-core transform color correction method yield better color correction results and are more robust. These approaches reduce the multicollinearity of the RGBN camera channels and will be a valuable reference in the development of RGBN imaging applications.
Most imaging devices lose image information during the acquisition process due to their low dynamic range (LDR). Existing high dynamic range (HDR) imaging techniques have a trade-off with time or spatial resolution, resulting in potential motion blur or image misalignment. Current HDR methods are based on the fusion of multi-frame LDR images and can suffer from blurring of fine details, image aliasing, and image boundary effects. This study developed a dual-channel camera (DCC) to achieve HDR imaging, which can eliminate image motion blur and registration problems. Considering the output characteristics of the camera, we propose a weighted sparse representation multi-scale transform fusion algorithm, which fully preserves the original image information, while eliminating image aliasing and boundary problems in the fused image, resulting in high-quality HDR imaging.
Generally, the dynamic range of night vision scenes is large. Owing to the limited dynamic range of traditional low light imaging technology, the captured images are always partially overexposed or underexposed. Multi-exposure fusion is the most effective method of overcoming the dynamic range limitation of sensor, and multi-frame low dynamic range (LDR) image fusion is a key consideration. However, existing fusion methods have problems such as image detail blurring and image aliasing. This paper proposes an image multi-scale decomposition method based on gradient domain guided filter (GDGF), which can better extract image details. The fusion algorithm adopts different fusion strategies for different scales. The low-frequency layer of the image uses a new weighted sparse representation (wSR) method, which can eliminate the image boundary problems and more adequately retain the image edges.
Hazardous gas leakage in the petrochemical industry frequently results in major incidents. A significant challenge arises due to the limitations of the current gas plume target feature extraction and identification techniques, which reduce the automated detection capabilities of remote monitoring systems. To address this, we propose BBGFA-YOLO, a real-time detection method leveraging background information and an improved YOLO network. This approach is designed specifically for the infrared imaging of gas plume targets, fulfilling the requirements of visual remote monitoring for hazardous gas leaks. We introduce a synthetic image colorization method based on background estimation, which leverages background estimation techniques to integrate motion features from gas plumes within the synthesized images. The resulting dataset can be directly employed by existing target detection networks. Furthermore, we introduce the MSDC-AEM, an attention enhancement module based on multi-scale deformable convolution, designed to enhance the network’s perception of gas plume features. Additionally, we incorporate an improved C2f-WTConv module, utilizing wavelet convolution, within the neck stage of the YOLO network. This modification strengthens the network’s capacity to learn deep gas plume features. Finally, to further optimize the network performance, we pre-train the network using a large-scale smoke detection dataset that includes reference background information. The experimental results, based on our self-acquired gas plume dataset, demonstrate a significant improvement in detection accuracy with the BBGFA-YOLO method, specifically achieving an increase in the average precision (AP50) from 74.2% to 96.2%. This research makes a substantial contribution to industrial hazardous gas leak detection technology, automated alarm systems, and the development of advanced monitoring equipment.
RGBN cameras that can capture visible light and near-infrared (NIR) light simultaneously produce better color image quality in low-light-level conditions. However, these RGBN cameras introduce additional color bias caused by the mixing of visible information and NIR information. The color correction matrix model widely used in current commercial color digital cameras cannot handle the complicated mapping function between biased color and ground truth color. Convolutional neural networks (CNNs) are good at fitting such complicated relationships, but they require a large quantity of training image pairs of different scenes. In order to achieve satisfactory training results, large amounts of data must be captured manually, even when data augmentation techniques are applied, requiring significant time and effort. Hence, a data generation method for training pairs that are consistent with target RGBN camera parameters, based on an open access RGB-NIR dataset, is proposed. The proposed method is verified by training an RGBN camera color restoration CNN model with generated data. The results show that the CNN model trained with the generated data can achieve satisfactory RGBN color restoration performance with different RGBN sensors.
Uncooled infrared imaging systems have significant potential in industrial hazardous gas leak detection. However, the use of narrowband filters to match gas spectral absorption peaks leads to a low level of incident energy captured by uncooled infrared cameras. This results in a mixture of fixed pattern noise and Gaussian noise, while existing denoising methods for uncooled infrared images struggle to effectively address this mixed noise, severely hindering the extraction and identification of actual gas leak plumes. This paper presents a UNet-structured dual-encoder denoising network specifically designed for narrowband uncooled infrared images. Based on the distinct characteristics of Gaussian random noise and row–column stripe noise, we developed a basic scale residual attention (BSRA) encoder and an enlarged scale residual attention (ESRA) encoder. These two encoder branches perform noise perception and encoding across different receptive fields, allowing for the fusion of noise features from both scales. The combined features are then input into the decoder for reconstruction, resulting in high-quality infrared images. Experimental results demonstrate that our method effectively denoises composite noise, achieving the best results according to both objective metrics and subjective evaluations. This research method significantly enhances the signal-to-noise ratio of narrowband uncooled infrared images, demonstrating substantial application potential in fields such as industrial hazardous gas detection, remote sensing imaging, and medical imaging.