Cloud detection is one of important tasks in automatic ground‐based cloud observation systems with ground‐based cloud images. Most supervised methods need substantial annotated samples for model training, while the pixel‐level annotation costs a lot. In this letter, a self‐training mechanism is proposed to significantly reduce the requirement of annotated samples. With a number of original images, only a few images need to be annotated (even incompletely), and a local region classifier model can be initialized with the annotated samples. Then the model is retrained iteratively using unlabeled samples with high confidence pseudo labels given by a fusion decision. The finely trained model can classify the local regions into “cloud” or “sky”. The experiments show that the proposed mechanism is effective for several classifiers. The proposed method can outperform unsupervised methods and achieve comparable results with fully supervised learning methods but using much fewer annotated samples.
21 June 2022 in IEEE Geoscience and Remote Sensing Letters
Cloud detection and recognition are two important tasks usually referring to image binary segmentation and image-level classification individually. Cloud genera segmentation has more practical significance but is much more challenging as a fine-grained pixel-level dense prediction problem. In this letter, a cascade cloud attribute discrimination network (CCAD-Net) is proposed. Based on an improved encoding-decoding model, CCAD-Net adds a binary segmentation branch for cloud detection and a attribute discrimination branch for cloud attribute feature learning in the decoding stage. Especially, in the attribute discrimination branch, several visual attributes are selected to design the attribute discrimination constraint according to prior professional knowledge and corresponding loss function is defined. These two additional branches and the final cloud genera segmentation branch extract their task-specific features successively and form a cascade structure. Due to the fusion of raw feature, binary segmentation feature, attribute discrimination feature, and cloud genera feature, CCAD-Net can achieve significant better performance than the state-of-the-art methods in cloud genera segmentation in whole-sky images.
1 January 2014 in Journal of Pattern Recognition Research
The Journal of Pattern Recognition Research (JPRR) provides an open access forum for the publication of research articles in areas of pattern recognition, machine learning, artificial intelligence, computational algorithms, and fuzzy learning
The automatic inspection of photovoltaic panels based on infrared images is one of the important tasks in the daily maintenance of photovoltaic panels in photovoltaic power plants. In this paper, a defect detection method of infrared thermal image photovoltaic panel based on morphological segmentation is proposed. First of all, according to the infrared characteristics of the photovoltaic plant station and the morphological characteristics of the photovoltaic panel, the accurate region of each photovoltaic panel is determined in the inspection image. Then, statistical method is used for feature analysis to determine whether there are defects in photovoltaic panels, and then the defective photovoltaic panels are segmented and different defects are classified according to morphological characteristics, and finally the defect detection and classification of photovoltaic panels are realized. Through the experiment in the actual inspection image of photovoltaic plant station, it is verified that the proposed method has a high accuracy of defect detection and recognition.
The fringe projection 3D measurement method is a widely used 3D measurement technology, which utilizes a digital projector as the structural light source. However, the nonlinear intensity response of projectors results in non-sinusoidal fringe images captured by cameras, introducing phase errors that undermine measurement accuracy. We propose a nonlinear correction method based on precise pre-coding for digital fringe projection 3D measurement systems. By encoding only two sets of four-step phase-shifting fringe images into the projector and modeling the relationship among gamma value, pre-coding value, and ideal phase, the algorithm estimates accurate pre-coding values. Experimental validation demonstrates significant reductions in phase error standard deviation post-gamma correction on standard whiteboards, leading to an ∼84% increase in 3D point cloud height accuracy. In addition, the proposed method effectively mitigates the periodic water ripple phenomenon induced by nonlinear gamma effects. Experimental investigations on plaster sculpture and printed circuit board further validate the method’s superiority, achieving higher phase accuracy and point cloud precision, with a measurement accuracy of 0.1 mm.