Challenges in image aesthetics assessment (IAA) arise from that images of different themes correspond to different evaluation criteria, and learning aesthetics directly from images while ignoring the impact of theme variations on human visual perception inhibits the further development of IAA; however, existing IAA datasets and models overlook this problem. To address this issue, we show that a theme-oriented dataset and model design are effective for IAA. Specifically, 1) we elaborately build a novel dataset, called TAD66K, that contains 66K images covering 47 popular themes, and each image is densely annotated by more than 1200 people with dedicated theme evaluation criteria. 2) We develop a baseline model, TANet, which can effectively extract theme information and adaptively establish perception rules to evaluate images with different themes. 3) We develop a large-scale benchmark (the most comprehensive thus far) by comparing 17 methods with TANet on three representative datasets: AVA, FLICKR-AES and the proposed TAD66K, TANet achieves state-of-the-art performance on all three datasets. Our work offers the community an opportunity to explore more challenging directions; the code, dataset and supplementary material are available at https://github.com/woshidandan/TANet.

Condition classification has been widely used for assessing equipment status for machine condition monitoring and diagnostics. An engine was fitted with one temperature and two pressure sensors to study the machine conditions in prognostics with an added abnormal state, in addition to the conventional normal and failure states. This work enables a better classification capability in order to predict deterioration in the engine. Information related to three deterioration processes was collected, and preprocessed using singular point elimination, deviation value acquisition, and data normalization. Wavelet transforms were used to extract deterioration features with different mother wavelets. The mother wavelets were selected using tests to optimize the wavelet selection. The deterioration was related to the amount of anomaly, with the abnormal states defined to distinguish the functional from the failure states. A Learning Vector Quantization (LVQ) neural network was used to classify the machine conditions, including normal, abnormal, and failure states. The results showed that the deterioration features defined using the Daubechies wavelet (db8) most strongly correlated with the original signal, so that the classification accuracy based on the deterioration features was greatly improved. The LVQ classification system had good accuracy for machine condition classification, and was adaptable to various engine conditions.