Chang Hong Lin

In vascular imaging, differences in near-infrared (NIR) absorption between intravascular hemoglobin and surrounding tissue modulate image intensity and contrast. Variations in illumination wavelength alter tissue absorption and scattering, thereby affecting optical penetration depth and vessel visibility. Consequently, image quality depends on both light source properties and imaging system geometry. To achieve uniform illumination and reduce skin surface reflection and scattering, this study employed NIR light-emitting diode (LED) ring light sources with wavelengths of 703 nm, 803 nm, and 901 nm. These NIR LED ring light sources were used with different optical packaging structures for vessel imaging experiments. A ring light module producing radially polarized illumination was fabricated by packaging the NIR LED with a reflective wire-grid polarization film to enhance polarization efficiency. Imaging results were observed and compared with those of two control groups (unpackaged and diffuser-packaged light sources). The beam was incident at Brewster’s angle to minimize specular reflection, and comparisons were performed using histogram-equalized images. The results show that when using a radially polarized ring light, the images of skin surface grooves are less pronounced, indicating that radial polarization can effectively reduce surface reflection from the skin.

This article proposes a scalable accelerator for deep learning (DL) implementation on edge computing, which is often limited by power, storage, and computation speed. The accelerator is based on systolic array cores with 126 processing elements (PEs) and optimized for YOLOv3-Tiny with $448\times448$ input images. Two multicast (MC) network architectures, feature map multicasting and weight multicasting, are introduced to control data stream distribution within the multicores. Results show that the proposed weight multicast (W-MC) systems outperformed the feature map multicast (FMAP-MC) systems in multicore scenarios, with up to $2.23\times $ frame rates per second (FPS). The 4-core W-MC system achieved the best efficiency with an overall frame rate of 13.73 FPS/W and an overall throughput of 35.83 GOPS/W. The 8-core W-MC system delivered the best performance, with a frame rate of 38.50 FPS after normalization to the standard YOLOv3-Tiny network. The proposed accelerator offers better computational efficiency and greater accelerator utilization in real-world inference scenarios, compared to previous state-of-the-art works.

As all kinds of self-contained display products have become popular, various technologies for improving printing applications have gradually been ignored or even stagnated. However, the number of printed matters in life has not decreased. Every day our lives are full of various printed materials, such as newspapers, magazines, leaflets, product packaging, and so on. The waste and pollution caused by printing technology are usually recessive, and not everyone values it. Unfortunately, people often overlook the waste of resources and environmental pollution caused by the manufacturing process, ink, toner, and printers. The method proposed in this paper uses a color difference model to find a manner that is not easy to notice by the human eye to save ink. The proposed method considers feasibility and practicability, and users can quickly find suitable settings for their applications. Also, through the actual printed pictures, the experimental results show that only 67.57% of the original ink consumption is required to obtain an equivalent print quality that is difficult to perceive by the human eye. Our proposed method will benefit the printing industry and reduce ink consumption on consumer printers, contributing to cost control and being nature-friendly.

As the techniques of autonomous driving become increasingly valued and universal, real-time semantic segmentation has become very popular and challenging in the field of deep learning and computer vision in recent years. However, in order to apply the deep learning model to edge devices accompanying sensors on vehicles, we need to design a structure that has the best trade-off between accuracy and inference time. In previous works, several methods sacrificed accuracy to obtain a faster inference time, while others aimed to find the best accuracy under the condition of real time. Nevertheless, the accuracies of previous real-time semantic segmentation methods still have a large gap compared to general semantic segmentation methods. As a result, we propose a network architecture based on a dual encoder and a self-attention mechanism. Compared with preceding works, we achieved a 78.6% mIoU with a speed of 39.4 FPS with a 1024 × 2048 resolution on a Cityscapes test submission.

Here, the feature occlusion, a data augmentation method that simulates real‐life challenges on the main features of the human face for age and gender recognition is proposed. Previous methods achieved promising results on constrained data sets with strict environmental settings, but the results on unconstrained data sets are still far from perfect. The proposed method adopted three simple occlusion techniques, blackout, random brightness, and blur, and each simulates a different kind of challenge that would be encountered in real‐world applications. A modified cross‐entropy loss that gives less penalty to the age predictions that land on the adjacent classes of the ground truth class is also proposed. The effectiveness of our proposed method is verified by implementing the augmentation method and modified cross‐entropy loss on two different convolution neural networks, the slightly modified AdienceNet and the slightly modified VGG16, to perform age and gender classification. The proposed augmentation system improves the age and gender classification accuracy of the slightly modified AdienceNet network by 6.62% and 6.53% on the Adience data set, respectively. The proposed augmentation system also improves the age and gender classification accuracy of the slightly modified VGG16 network by 6.20% and 6.31% on the Adience data set, respectively.