In this paper, we propose a novel element-aware domain enhancement and adaptation (EDEA) approach for semantic segmentation to increase the segmentation accuracy. In the proposed EDEA approach, we first analyze the warning elements in the testing step, such as the falling-leaves, manhole covers, cirrus, advertisements, etc., caused invalid segmented objects. Then, we create a new GTA5-like (Grand Theft Auto V-like) dataset containing the scenarios including these warning elements. Further, we perform a domain adaptation on the created GTA5-like dataset to generate a photo-realistic GTA5-like dataset. Finally, we combine the generated dataset with the original photo-realistic GTA5 dataset and the realistic Camvid dataset to constitute a more diverse training dataset. The comprehensive experimental results have confirmed the semantic segmentation accuracy improvement of the proposed EDEA approach relative to the previous two domain adaptation methods.
Binarizing historically degraded as-built drawing (HDAD) maps is a challenging job, especially in removing noise, yellowing areas, and folded lines while preserving the foreground components. This paper first proposes a convolutional neural networks-based (CNN-based) color classifier to determine the dominant color class of each HDAD block. Then, a dominant color driven- and CNN-based binarization method is proposed, producing a high-quality binarized HDAD map. Based on real HDAD dataset, the thorough experiments have been carried out to show that in terms of F-measure and perceptual effect, our binarization method substantially outperforms existing state-of-the art binarization methods.
The Bayer color filter array (CFA) pattern is the most widely used CFA pattern in the digital color cameras market. The chroma 4:2:0 subsampling of Bayer CFA images is a necessary process prior to compression. In this paper, based on the CFA block-distortion minimization criterion, we propose an effective region-based chroma 4:2:0 subsampling method for Bayer CFA images. Based on the test Kodak and IMAX datasets, the experimental results demonstrated that in the current high efficiency video coding (HEVC) reference software HM-16.18, our method has substantial quality of the reconstructed images when compared with the existing six chroma subsampling methods.
In this paper, we propose a novel Vision- and system on chip (SoC)- based fall detection method for the elderly. Once, a fall event is detected, an alarm signal is immediately sent out to query first aid to the elderly. Our novel fall detection method consists of five effective steps: checking whether the light condition has been stabilized, GMM-based background and foreground estimation, a new strategy to solve the foreground lag problem, solving the false fall detection problem when light comes from a neighboring room, as well as the fall detection determination and the general-purpose input/output based warning mechanism. Based on the test videos, the experiments have been carried out demonstrate that our proposed fall detection method can meet the real-time, low cost, and high accuracy demands.