Language：Japanese   Publisher：Japan Society of Material Cycles and Waste Management  

<p>In this study, an AI technology called YOLO that is a popular object detection and image segmentation model was used to construct a system to support recovered paper separating and its accuracy was verified. The system inputs images of 12 types of paper products and uses AI (YOLO) to detect their location and identify their types. The data used were a total of 6,000 images for training and 1,500 images with multiple types of paper products on each image for verification. As a result, the percentage of images that could be detected and discriminated (mAP) was 71.4%. When images with overlapping paper products were removed, the mAP was 83.6%. In order to improve detection accuracy in the future, it was mentioned that learning more cases with overlapping paper products and more types of angles at which photos are taken should be used.</p>

DOI： 10.14912/jsmcwm.35.0_7

CiNii Research

Language：Japanese   Publisher：Japan Society of Material Cycles and Waste Management  

<p>Regarding waste paper collection in Japan, there is room for improvement in the level of awareness and implementation of the regulations, and efforts are needed to encourage people to put them into practice. This research aims to use a method called nudge and set up posters to bring about behavioral changes that encourage people to collect waste paper. Nudge is a concept from behavioral economics that refers to a method of guiding people to make better choices without restricting or coercing them.Posters about environmental risks and simple informational purposes were placed on the Ito Campus of Kyushu University at the Faculties of Engineering and Agriculture, respectively. The study assessed the amount of waste paper and its proportion in burnable garbage, finding no significant change. Despite 57.2% of Engineering students and 40% of Agriculture students understanding the importance of waste classification, behavior did not notably improve.</p>

DOI： 10.14912/jsmcwm.35.0_89

CiNii Research

Language：Japanese   Publisher：Japan Society of Material Cycles and Waste Management  

<p>Clothing production has increased significantly worldwide, and the amount of clothing discarded has increased accordingly. However, one of the reasons for the lack of progress in clothing reuse is the low level of awareness of clothing reuse. Therefore, research was conducted on ways to improve consumer awareness. Specifically, the research aimed to increase consumer awareness of recycling by using IoT tags to visualize the condition of clothing and messages from previous owners. Therefore, we conducted a questionnaire survey and an experiment with students using the tags. Through the questionnaire survey and the experiment, a radar chart visualizing the condition of used clothing (dirt, wrinkles, etc.) and messages from the previous owner to the buyer of used clothing were considered effective. On the other hand, it was pointed out that the amount of information on the IoT tags was insufficient in this experiment, and it is necessary to increase the amount of useful information through future surveys and experiments.</p>

DOI： 10.14912/jsmcwm.35.0_265

CiNii Research

Language：Japanese   Publisher：Japan Society of Material Cycles and Waste Management  

<p>Recycling paper is an important issue in waste reduction and decarbonization. Fukuoka City has been working on the sorting of used paper as part of its waste reduction policy, but small and medium-sized enterprises (SMEs) are less likely to sort used paper than large enterprises (LSEs). In this study, with the aim of improving the sorting rate by reducing the burden of sorting, we developed a prototype system that automatically supports sorting of paper using image recognition technology based on deep learning called CNN, and examined its construction. The results showed that the use of a large-scale pre-trained model with deep layers resulted in an 88% correct inference rate by image recognition. In addition, it was found that the correct inference rate for the major categories as burnable trash and used paper was over 90% for each.</p>

DOI： 10.14912/jsmcwm.34.0_23

CiNii Research

Language：Japanese   Publisher：Japan Society of Material Cycles and Waste Management  

<p>Currently, the earth is facing problems such as resource depletion and global warming. One way to solve these problems is to centrally manage and utilize a series of processes from product manufacturing, consumption, waste discharge/collection, intermediate treatment, recycling, and final disposal on a platform to process waste. and optimization of resource recycling. This study clarifies the relationship between the retail sales volume, which is the artery of this platform, and the general waste volume, which is the vein, and predicts the waste discharge volume using retail sales data.</p>

DOI： 10.14912/jsmcwm.34.0_315

CiNii Research