Recognition of Similar Weather Scenarios in Terminal Area Based on Contrastive Learning

In order to improve the recognition accuracy of similar weather scenarios(SWSs)in terminal area,a recognition model for SWS based on contrastive learning(SWS-CL)is proposed.Firstly,a data augmentation method is designed to improve the number and quality of weather scenarios samples according to the characteristics of convective weather images.Secondly,in the pre-trained recognition model of SWS-CL,a loss function is formulated to minimize the distance between the anchor and positive samples,and maximize the distance between the anchor and the negative samples in the latent space.Finally,the pre-trained SWS-CL model is fine-tuned with labeled samples to improve the recognition accuracy of SWS.The comparative experiments on the weather images of Guangzhou terminal area show that the proposed data augmentation method can effectively improve the quality of weather image dataset,and the proposed SWS-CL model can achieve satisfactory recognition accuracy.It is also verified that the fine-tuned SWS-CL model has obvious advantages in datasets with sparse labels.

RESEARCH ON COLOR CONSTANCY UNDER OPEN ILLUMINATION CONDITIONS

Color information plays a key role in the research fields of object recognition and image retrieval. However, the actual color varies by the conditions of illumination, especially the open natural daylight. Four different color constancy schemes are proposed in the paper to minimize the effects of open illumination conditions. (1) The color constancy scheme based on the image statistics is proposed, which includes the color cast detection and removal. (2) The color constancy scheme based on the color temperature curve is proposed, which combines Gaussian model with linear fitting to estimate color temperature curve. (3) The color constancy scheme based on the double exposure theory is proposed, which is able to reproduce a color image under typical illumination. (4) According to the concepts of supervised learning, the supervised color constancy scheme is proposed. The transformation of color values from unknown illumination to typical illumination is solved by improved Support Vector Regression (SVR).

A joint matrix minimization approach for multi-image face recognition

The Schatten p-quasi-norm regularized minimization problem has attracted extensive attention in machine learning, image recognition, signal reconstruction, etc. Meanwhile, the l_(2,1)-regularized matrix optimization models are also popularly used for its joint sparsity. Naturally, the pseudo matrix norm l_(2,p) is expected to carry over the advantages of both l_p and l_(2,1). This paper proposes a mixed l_(2,q)-l_(2,p) matrix minimization approach for multi-image face recognition. To uniformly solve this optimization problem for any q ∈ [1,2] and p ∈(0,2], an iterative quadratic method(IQM) is developed. IQM is proved to iescend strictly until it gets a stationary point of the mixed l_(2,q)-l_(2,p)matrix minimization. Moreover, a more practical IQM is presented for large-scale case. Experimental results on three public facial image databases show that the joint matrix minimization approach with practical IQM not only saves much computational cost but also achievez better performance in face recognition than state-of-the-art methods.

Affective rating ranking based on face images in arousal-valence dimensional space

In dimensional affect recognition, the machine learning methods, which are used to model and predict affect, are mostly classification and regression. However, the annotation in the dimensional affect space usually takes the form of a continuous real value which has an ordinal property. The aforementioned methods do not focus on taking advantage of this important information. Therefore, we propose an affective rating ranking framework for affect recognition based on face images in the valence and arousal dimensional space. Our approach can appropriately use the ordinal information among affective ratings which are generated by discretizing continuous annotations.Specifically, we first train a series of basic cost-sensitive binary classifiers, each of which uses all samples relabeled according to the comparison results between corresponding ratings and a given rank of a binary classifier. We obtain the final affective ratings by aggregating the outputs of binary classifiers. By comparing the experimental results with the baseline and deep learning based classification and regression methods on the benchmarking database of the AVEC 2015 Challenge and the selected subset of SEMAINE database, we find that our ordinal ranking method is effective in both arousal and valence dimensions.