In this study, we propose a framework to solve the cable traction problem where many cables are needed, such as in factories, schools, etc. Our framework is divided into three stages: First, we use ...SegNet to recognize the cable in the image, build a 3D voxel model by 3D-VAE-GAN, and transform the voxel model to a 3D particle model in Unity3D by the Position-Based Dynamics method. Second, features learned by a 3D deep neural network (PointNet++) from the 3D model are fed into a deep reinforcement learning (RL) network (DQN) to get a series of actions to untie the 3D cable model in Unity3D. Finally, we input the action sequence to the robot arm to untie the cable in the real world. Experimental results show that compared with traditional methods, our method has successfully applied artificial intelligence algorithms to help the computer learn how to untie knotted cables by itself in the virtual world. The untying operations learned in the virtual world can be also used to untie the cable in the real world.
Image recognition models that work in challenging environments (e.g., extremely dark, blurry, or high dynamic range conditions) must be useful. However, creating training datasets for such ...environments is expensive and hard due to the difficulties of data collection and annotation. It is desirable if we could get a robust model without the need for hard-to-obtain datasets. One simple approach is to apply data augmentation such as color jitter and blur to standard RGB (sRGB) images in simple scenes. Unfortunately, this approach struggles to yield realistic images in terms of pixel intensity and noise distribution due to not considering the non-linearity of Image Signal Processors (ISPs) and noise characteristics of image sensors. Instead, we propose a noise-accounted RAW image augmentation method. In essence, color jitter and blur augmentation are applied to a RAW image before applying non-linear ISP, resulting in realistic intensity. Furthermore, we introduce a noise amount alignment method that calibrates the domain gap in the noise property caused by the augmentation. We show that our proposed noise-accounted RAW augmentation method doubles the image recognition accuracy in challenging environments only with simple training data.
In order to establish a real-time measurement system for the concentration of airborne pollen allergens, we examined Cry j 1, one of the major allergens of Cryptomeria japonica pollen, as an example ...to establish the system. The feasible system consisted of: collection of airborne pollen allergens using the Virtual Impactor or Cyclone sampler, extraction of Cry j 1 using 10 mM HEPES buffer containing 0.125 M NH sub(4)HCO sub(3) and finally, real-time measurement of Cry j 1 using the BIACORE 3000 system. The sensitivity of the system was 5 ng/ml, and 0.1 ml sample volume and at least 500 pg of Cry j 1 were required for each measurement. Quantification of Cry j 1 in the air can be determined 30 min after collection, i.e. 15 min for extraction, 10 min for separation from particulate matters and 5 min for the measurement.
This book constitutes the proceedings of the 11th RoboCup International Symposium, held in Atlanta, GA, USA, in July 2007, immediately after the 2007 RoboCupSoccer, RoboCupRescue and RoboCupJunior ...competitions. Papers presented at the symposium focused on topics related to these three events and to artificial intelligence and robotics in general. The 18 revised full papers and 42 revised poster papers included in the book were selected from 133 submissions. Each paper was reviewed by at least three program committee members. The program committee also nominated two papers for the Best Paper and Best Student Paper awards, respectively. The book provides a valuable source of reference and inspiration for R&D professionals and educationalists active or interested in robotics and artificial intelligence.
Monodisperse, submicron-sized, reactive polystyrene (PST) microspheres having active ester groups on their surfaces were prepared by emulsifier-free emulsion co-polymerizations of ST with ...water-soluble, active ester monomers (AEMs) which were newly synthesized. The resulting microspheres showed characteristics of a high density of active ester groups on their surfaces and a high reactivity towards primary amines. Some applications using the highly reactive microspheres were attempted in the preparation of supraparticles and in the surface modification of solids.
In the field of systems biology, biochemical networks are being reconstructed in computer to understand their dynamic features. In this study, we focused on the estimation problem for a dynamic model ...of the Drosophila circadian oscillator, which is defined by two evaluation functions that represent oscillatory features. However, since the search space is multimodal and logarithmically large, it is quite difficult for ordinary GAs to optimize this evaluation problem. Therefore, we had used two-step optimizing, a random search with GA. It successfully optimized the circadian oscillator, but required a long calculation time, causing a local search. On the other hand, to optimize two evaluation functions simultaneously, we proposed the survival ratio GA, where genes have lifetimes and a population holds search histories. By alternating two evaluation functions every generation, the population holds previously evaluated genes and children inherit each and mixed properties. The survival ratio GA exhibited wider space search and higher success ratio, thus we applied the one-step optimizing with the survival ratio GA to the circadian estimation problem, which shortens the total calculation times and finds out more local optimums than the previous two-step optimizing method.
Full DNN-based image signal processors (ISPs) have been actively studied and have achieved superior image quality compared to conventional ISPs. In contrast to this trend, we propose a lightweight ...ISP that consists of simple conventional ISP functions but achieves high image quality by increasing expressiveness. Specifically, instead of tuning the parameters of the ISP, we propose to control them dynamically for each environment and even locally. As a result, state-of-the-art accuracy is achieved on various datasets, including other tasks like tone mapping and image enhancement, even though ours is lighter than DNN-based ISPs. Additionally, our method can process different image sensors with a single ISP through dynamic control, whereas conventional methods require training for each sensor.
Unprocessed sensor outputs (RAW images) potentially improve both low-level and high-level computer vision algorithms, but the lack of large-scale RAW image datasets is a barrier to research. Thus, ...reversed Image Signal Processing (ISP) which converts existing RGB images into RAW images has been studied. However, most existing methods require camera-specific metadata or paired RGB and RAW images to model the conversion, and they are not always available. In addition, there are issues in handling diverse ISPs and recovering global illumination. To tackle these limitations, we propose a self-supervised reversed ISP method that does not require metadata and paired images. The proposed method converts a RGB image into a RAW-like image taken in the same environment with the same sensor as a reference RAW image by dynamically selecting parameters of the reversed ISP pipeline based on the reference RAW image. The parameter selection is trained via pseudo paired data created from unpaired RGB and RAW images. We show that the proposed method is able to learn various reversed ISPs with comparable accuracy to other state-of-the-art supervised methods and convert unknown RGB images from COCO and Flickr1M to target RAW-like images more accurately in terms of pixel distribution. We also demonstrate that our generated RAW images improve performance on real RAW image object detection task.
