Automatic segmentation of retinal images is an important task in computer-assisted medical image analysis for the diagnosis of diseases such as hypertension, diabetic and hypertensive retinopathy, and arteriosclerosis. Among the diseases, diabetic retinopathy, which is the leading cause of vision detachment, can be diagnosed early through the detection of retinal vessels. The manual detection of these retinal vessels is a time-consuming process that can be automated with the help of artificial intelligence with deep learning. The detection of vessels is difficult due to intensity variation and noise from non-ideal imaging. Although there are deep learning approaches for vessel segmentation, these methods require many trainable parameters, which increase the network complexity. To address these issues, this paper presents a dual-residual-stream-based vessel segmentation network (Vess-Net), which is not as deep as conventional semantic segmentation networks, but provides good segmentation with few trainable parameters and layers. The method takes advantage of artificial intelligence for semantic segmentation to aid the diagnosis of retinopathy. To evaluate the proposed Vess-Net method, experiments were conducted with three publicly available datasets for vessel segmentation: digital retinal images for vessel extraction (DRIVE), the Child Heart Health Study in England (CHASE-DB1), and structured analysis of retina (STARE). Experimental results show that Vess-Net achieved superior performance for all datasets with sensitivity (Se), specificity (Sp), area under the curve (AUC), and accuracy (Acc) of 80.22%, 98.1%, 98.2%, and 96.55% for DRVIE; 82.06%, 98.41%, 98.0%, and 97.26% for CHASE-DB1; and 85.26%, 97.91%, 98.83%, and 96.97% for STARE dataset.

Recently, autonomous vehicles, particularly self-driving cars, have received significant attention owing to rapid advancements in sensor and computation technologies. In addition to traffic sign recognition, road lane detection is one of the most important factors used in lane departure warning systems and autonomous vehicles for maintaining the safety of semi-autonomous and fully autonomous systems. Unlike traffic signs, road lanes are easily damaged by both internal and external factors such as road quality, occlusion (traffic on the road), weather conditions, and illumination (shadows from objects such as cars, trees, and buildings). Obtaining clear road lane markings for recognition processing is a difficult challenge. Therefore, we propose a method to overcome various illumination problems, particularly severe shadows, by using fuzzy system and line segment detector algorithms to obtain better results for detecting road lanes by a visible light camera sensor. Experimental results from three open databases, Caltech dataset, Santiago Lanes dataset (SLD), and Road Marking dataset, showed that our method outperformed conventional lane detection methods.

In recent years, the importance of semantic segmentation has been widely recognized and the field has been actively studied. The existing state-of-the-art segmentation methods show high performance for bright and clear images. However, in low light or nighttime environments, images are blurred and noise increases due to the nature of the camera sensor, which makes it very difficult to perform segmentation for various objects. For this reason, there are few previous studies on multi-class segmentation in low light or nighttime environments. To address this challenge, we propose a modified cycle generative adversarial network (CycleGAN)-based multi-class segmentation method that improves multi-class segmentation performance for low light images. In this study, we used low light databases generated by two road scene open databases that provide segmentation labels, which are the Cambridge-driving labeled video database (CamVid) and Karlsruhe Institute of Technology and Toyota Technological Institute at Chicago (KITTI) database. Consequently, the proposed method showed superior segmentation performance compared with the other state-of-the-art methods.

Conventional nighttime face detection studies mostly use near-infrared (NIR) light cameras or thermal cameras, which are robust to environmental illumination variation and low illumination. However, for the NIR camera, it is difficult to adjust the intensity and angle of the additional NIR illuminator according to its distance from an object. As for the thermal camera, it is expensive to use as a surveillance camera. For these reasons, we propose a nighttime face detection method based on deep learning using a single visible-light camera. In a long-distance night image, it is difficult to detect faces directly from the entire image due to noise and image blur. Therefore, we propose Two-Step Faster region-based convolutional neural network (R-CNN) based on the image preprocessed by histogram equalization (HE). As a two-step scheme, our method sequentially performs the detectors of body and face areas, and locates the face inside a limited body area. By using our two-step method, the processing time by Faster R-CNN can be reduced while maintaining the accuracy of face detection by Faster R-CNN. Using a self-constructed database called Dongguk Nighttime Face Detection database (DNFD-DB1) and an open database of Fudan University, we proved that the proposed method performs better compared to other existing face detectors. In addition, the proposed Two-Step Faster R-CNN outperformed single Faster R-CNN and our method with HE showed higher accuracies than those without our preprocessing in nighttime face detection.

Owing to the rapid advancements in smartphone technology, there is an emerging need for a technology that can detect banknotes and coins to assist visually impaired people using the cameras embedded in smartphones. Previous studies have mostly used handcrafted feature-based methods, such as scaleinvariant feature transform or speeded-up robust features, which cannot produce robust detection results for banknotes or coins captured in various backgrounds and environments. With the recent advancement in deep learning technology, some studies have been conducted on banknote and coin detection using a deep convolutional neural network (CNN). However, these studies also showed degraded performance depending on the changes in background and environment. To overcome these drawbacks, this paper proposes a threestage detection technology for new banknotes and coins by applying faster region-based CNN, geometric constraints, and the residual network (ResNet). In the experiment performed using the open database of Jordanian dinar (JOD) and 6,400 images of eight types of Korean won banknotes and coins obtained using our smartphones, the proposed method exhibited a better detection performance than the state-of-the-art methods based on handcrafted features and deep features.