This paper proposes facial recognition approach on variations in outdoor lighting conditions using short wavelength infrared (SWIR) images. Conventional visible light (VIS) systems struggle under challenging lighting, such as backlit condition, due to their sensitivity to ambient light. Recently, SWIR imaging has emerged as a promising alternative, offering robustness against such conditions. Despite its advantages, SWIR imaging faces challenges, including higher noise levels, limited high-intensity illumination, and differences in image characteristics compared to VIS images. These challenges hinder the integration of SWIR-based systems with existing VIS databases. To address these issues, this study proposes a SWIR-to-VIS image translation method to improve VIS-SWIR face recognition performance. The proposed method relies solely on recognition loss for training the image translator, eliminating the need for other loss functions. This novel approach ensures compatibility with existing VIS-based databases while taking advantage of SWIR’s robustness to ambient light. Experimental evaluations demonstrate the efficacy of the proposed method in achieving accurate face recognition under severe outdoor lighting conditions.

Low-Light Image Enhancement (LLIE) is crucial for improving both human perception and computer vision tasks. This paper addresses two challenges in zero-reference LLIE: obtaining perceptually ’good’ images using the Contrastive Language-Image Pre-Training (CLIP) model and maintaining computational efficiency for high-resolution images. We propose CLIP-Utilized Reinforcement learning-based Visual image Enhancement (CURVE). CURVE employs a simple image processing module which adjusts global image tone based on Bézier curve and estimates its processing parameters iteratively. The estimator is trained by reinforcement learning with rewards designed using CLIP text embeddings. Experiments on low-light and multi-exposure datasets demonstrate the performance of CURVE in terms of enhancement quality and processing speed compared to conventional methods.

This paper rethinks image histogram matching (HM) and proposes a differentiable and parametric HM preprocessing for a downstream classifier. Convolutional neural networks have demonstrated remarkable achievements in classification tasks. However, they often exhibit degraded performance on low-contrast images captured under adverse weather conditions. To maintain classifier performance under low-contrast images, histogram equalization (HE) is commonly used. HE is a special case of HM using a uniform distribution as a target pixel value distribution. In this paper, we focus on the shape of the target pixel value distribution. Compared to a uniform distribution, a single, well-designed distribution could have potential to improve the performance of the downstream classifier across various adverse weather conditions. Based on this hypothesis, we propose a differentiable and parametric HM that optimizes the target distribution using the loss function of the downstream classifier. This method addresses pixel value imbalances by transforming input images with arbitrary distributions into a target distribution optimized for the classifier. Our HM is trained on only normal weather images using the classifier. Experimental results show that a classifier trained with our proposed HM outperforms conventional preprocessing methods under adverse weather conditions.

We propose a target driven adaptive (TDA) loss to enhance the performance of infrared small target detection (IRSTD). Prior works have used loss functions, such as binary cross-entropy loss and IoU loss, to train segmentation models for IRSTD. Minimizing these loss functions guides models to extract pixel-level features or global image context. However, they have two issues: improving detection performance for local regions around the targets and enhancing robustness to small scale and low local contrast. To address these issues, the proposed TDA loss introduces a patch-based mechanism, and an adaptive adjustment strategy to scale and local contrast. The proposed TDA loss leads the model to focus on local regions around the targets and pay particular attention to targets with smaller scales and lower local contrast. We evaluate the proposed method on three datasets for IRSTD. The results demonstrate that the proposed TDA loss achieves better detection performance than existing losses on these datasets.

We propose an image-adaptive object detection method for adverse weather conditions such as fog and low-light. Our framework employs differentiable preprocessing filters to perform image enhancement suitable for later-stage object detections. Our framework introduces two differentiable filters: a Bezier curve-based pixel-wise (BPW) filter and a kernel-based local (KBL) filter. These filters unify the functions of classical image processing filters and improve performance of object detection. We also propose a domain-agnostic data augmentation strategy using the BPW filter. Our method does not require data-specific customization of the filter combinations, parameter ranges, and data augmentation. We evaluate our proposed approach, called Enhanced Robustness by Unified Image Processing (ERUP)-YOLO, by applying it to the YOLOv3 detector. Experiments on adverse weather datasets demonstrate that our proposed filters match or exceed the expressiveness of conventional methods and our ERUP-YOLO achieved superior performance in a wide range of adverse weather conditions, including fog and low-light conditions.

This paper proposes a deep feature extractor for iris recognition at arbitrary resolutions. Resolution degradation reduces the recognition performance of deep learning models trained by high-resolution images. Using various-resolution images for training can improve the model’s robustness while sacrificing recognition performance for high-resolution images. To achieve higher recognition performance at various resolutions, we propose a method of resolution-adaptive feature extraction with automatically switching networks. Our framework includes resolution expert modules specialized for different resolution degradations, including down-sampling and out-of-focus blurring. The framework automatically switches them depending on the degradation condition of an input image. Lower-resolution experts are trained by knowledge-distillation from the high-resolution expert in such a manner that both experts can extract common identity features. We applied our framework to three conventional neural network models. The experimental results show that our method enhances the recognition performance at low- resolution in the conventional methods and also maintains their performance at high-resolution.

With the development of a mobility society, supporting drivers from the mental aspect for safe driving is an increasingly important issue before self-driving cars are dominant. We have developed a sensor network that uses biosignal sensors such as EEG, ECG, heart rate, and drivers’ operation of pedals and steering to measure drivers’ emotions. However, even though EEG may be an excellent index to estimate emotion, it is not practical to wear an EEG sensor while driving. So, we would like to add other methods that are easy to use and can support estimating drivers’ emotions. In this paper, we report the result of using the facial expression analysis technique to estimate a car driver’s stress and fatigue. The result shows that facial expression reflects driver emotion in most cases and is closely related to automotive operating status.

User identification from hand images only is still a challenging task. In this paper, we propose a new biometric identification system based solely on a skin patch from a multispectral image. The system is utilizing a novel modified 3D CNN architecture which is taking advantage of multispectral data. We demonstrate the application of our system for the example of human identification from multispectral images of hands. To the best of our knowledge, this paper is the first to describe a pose-invariant and robust to overlapping real-time human identification system using hands. Additionally, we provide a framework to optimize the required spectral bands for the given spatial resolution limitations.

Manga (Japanese style comics) is one of the most popular publications. Nowadays manga is often handled as digital images not only in consumers' use but also in digital media. However, they hardly handle manga as content-based materials. Some digital media use tags or text data for retrieval, where the tags and text data are produced by handmade input. Therefore our goal is achieving content-based retrieval method for manga images. As the first step to the goal, we investigate the performance of Sun's method applying to manga character retrieval. Manga character retrieval means a image retrieval of which the input and output are a character image and page images where the input character appears respectively. It is useful for convenient use of manga images, for example, character retrieval or auto-tagging. We modify Sun's method so as to be applicable to manga character retrieval and then investigate the performance.

It is impossible to observe the Global Positioning System (GPS) signal traveling in Line of Sight (LOS) indoors. When the position is estimated by using Non-LOS GPS received signals, it results in serious positioning error because the conventional estimation method by the GPS supposes the GPS signals in LOS. However, we may observe reflected signals from the GPS satellites even if indoors. Thus, we consider that the reflected received signals are useful for a reference of the position estimation although it is impossible to directly estimate the position. Supposing that the Received Signal Strength Indicator (RSSI) of the GPS signal varies between locations, we create a database of the RSSI of the GPS signal in each location such as the well-known fingerprinting method. It is possible to estimate the user location by querying the database even if indoors. By experiments, we evaluate the proposed method and propose a shrink method for the database additionally. We show that the proposed method can estimate indoor location well enough.

A traditional camera requires the photographer to select the many parameters at capture time. While advances in light field photography have enabled post-capture control of focus and perspective, they suffer from several limitations including lower spatial resolution, need for hardware modifications, and restrictive choice of aperture and focus setting. In this paper, we propose "compressive epsilon photography," a technique for achieving complete post-capture control of focus and aperture in a traditional camera by acquiring a carefully selected set of 8 to 16 images and computationally reconstructing images corresponding to all other focus-aperture settings. We make the following contributions: first, we learn the statistical redundancies in focal-aperture stacks using a Gaussian Mixture Model; second, we derive a greedy sampling strategy for selecting the best focus-aperture settings; and third, we develop an algorithm for reconstructing the entire focal-aperture stack from a few captured images. As a consequence, only a burst of images with carefully selected camera settings are acquired. Post-capture, the user can then select any focal-aperture setting of choice and the corresponding image can be rendered using our algorithm. We show extensive results on several real data sets.

In this paper, we present an outline of the new Information Delivery System During a Major Disaster for People Who are Deaf (IDDD) designed using a web platform such as node.js and GCM (Google Cloud Message), and explain the performance measurement results. Especially, a new implementation method using web system for embedded system and M2M system is explained. Also, we explained delay to provide disaster information to a user and spreading that information to multiple displays are shorter than previous IDDD.

In 2007, the Ministry of Internal Affairs and Communications of Japan tested 16 different models of a safety support system for children on school routes. One of the models was constructed and tested at a school in an area of the city of Hiroshima from September to December of 2007. For the model project, we developed a new safety support system for children on school routes by using a mobile ad hoc network constructed from mobile phones with the Bluetooth function. The support system provided good performance and accuracy in maintaining the safety of students on the way to school. The basic idea of the safety support system is the grouping of children and volunteers using a mobile ad hoc network. In this paper, we present an outline of this system and evaluate the performance of grouping and the effectiveness of our approach.

Hiroshima City Children Tracking System is a safety support system for children based on ad hoc network technologies. Field experiments have been conducted in cooperation with an elementary school in Hiroshima. In this paper, we propose a new generation children tracking system which is based on experiences and findings of the field experiments for Hiroshima City Children Tracking System. Our proposed system consists of Android terminals which has Wireless LAN device and Bluetooth device with the ad hoc communication function. Our system manages groups of Android terminals using Autonomous Clustering technique. In this paper, we show the system requirements for our children tracking system and describe the implementation features to satisfy the system requirements. Finally, we provide some preliminary implemented results for our proposed system.

To keep safety is one of the most important duties of government. In 2007, Ministry of Internal Affairs and Communications of Japan tested 16 different models of security system for children on school route. One of the models was constructed and tested at an area of a school in Hiroshima City from September to December of 2007. A consortium was established by Hiroshima City, Hiroshima City University, Chugoku Electric Power Co., Inc, KDDI Corporation, etc. to conduct this project. For this model project, we developed new security system for children on school route by using mobile ad hoc network that is constructed based on mobile phone with Bluetooth function. About 700 students used this system for four months. In this paper, we present the outline of this system and the result of the trial. We also describe that this system has good performance and accuracy to keep safety of students on the way to school.

We present our trial of citizen's participation art as autonomous and decentralized System. Our purposes are to provide an environment to encourage people to join an event through ICT tools, and feel happiness and unity in a community.

The Software Development Project Scheduling Problem is similar to the well-known Resource-Constrained Multi-Project Scheduling Problem (RCMPSP). It consists in determining a schedule of tasks taking into consideration resource availabilities and precedence constraints, while optimizing an objective. Like RCMPSP, it is an NP-hard problem. In this paper, a task segmentation scheme to schedule a software development project is proposed and the average duration of the multiple concurrent projects is minimized using the Particle Swarm Optimization (PSO) meta-heuristic. PSO is a recent meta-heuristic algorithm, known for its simplicity in programming and its rapid convergence. A series of experiments show optimum results for several software development schedule scenarios.

We propose a new room-sized immersive projection display. The display consists of a cylindrical screen that can be moved horizontally and vertically, allowing the user to easily change his/her field of view by moving the screen to any angle. The angle of the screen is measured by a motion sensor, and the projected stereo images are changed in response to the measured angle.

There are about 90 million high performance mobile phones used in Japan. We are now planning to develop new applications of mobile phone to support children and elder and disabled people who are out of scope of major mobile phone application based on their requirements. We have a responsibility to extend the application filed of mobile phone as a leading country of ubiquitous life. This paper discusses possibilities to realize mobile ad hoc networks using Bluetooth functions equipped on a mobile phone. Hierarchical mobile ad hoc networks using Bluetooth in a mobile phone are firstly developed as a test platform. The test platform proves the possibility of developing mobile ad hoc network by mobile phone built-in Bluetooth functions. We demonstrate their capabilities by showing results of implementing game applications on the test platform. The paper also describes some example applications using mobile ad hoc network technologies, which include a location tracking system for children on the way to a school and an alarm system for hearing impaired people

Telecommunications are powerful tools to reduce loss of life and property damage during disasters. Especially, for the people with disabilities, information provided through their own telecommunication devices is indispensable to move to a place of refuge safely and rapidly. In this paper, we propose the wireless system to send information on disasters to people with disabilities (especially deaf people) and to identify their location at the refuge when they are evacuated. The system consists of cellular phone, web server and electronic display, which are concatenated by Short Message Service (SMS) and Bluetooth. By using these devices and public network services, the system provides a function to send appropriate information to people with disabilities rapidly. The system provides the network platform to share information about their whereabouts with related persons. Finally, we summarize results from demonstration of our system to the deaf people experienced large earthquake.

File-sharing Peer-to-Peer systems are effective for autonomous information retrieval over the networks. However, the previous information retrieval schemes such as Gnutella and Local Indices have bad performance and large overhead. In order to solve these drawbacks, this paper proposes a dynamic information retrieval scheme, in which indices are dynamically allocated in appropriate nodes adaptively to variation of traffic patterns caused by query messages. The simulation experimental results show that the proposed scheme has good performance with reasonable overhead even when the traffic patterns vary as time proceeds.