"Kabepita," an AI wallpaper identification app equipped with an "automatic texture identification program" jointly developed by the Intelligent Mechanism Laboratory of the Department of Intelligent Information Engineering, Faculty of Science and Engineering, Doshisha University, and Komatsu Corporation, won the Grand Prix at the "KANSAI DX AWARD 2024" (sponsored by the Kansai Digital Month Executive Committee). The award recognizes companies that are particularly outstanding as advanced examples of DX initiatives, and the awards ceremony was held at the Kansai Digital Month closing event on October 2024, 10.
The interior construction industry faces major challenges in terms of labor shortages and long working hours. There are many inefficient tasks that have been carried out over many years, one of which is the "identification of materials with unknown part numbers," which is done manually several times a month at the request of customers. Manual identification is inefficient, prone to errors, and labor-intensive, but it has been left unattended for a long time due to the lack of a technical solution.
Therefore, the Intelligent Mechanisms Laboratory of the Department of Intelligent Information Engineering, Faculty of Science and Engineering, Doshisha University, and Komatsu Corporation, an interior design company based in Higashi-Osaka with over 45 years of experience, set out to solve this problem using image recognition AI through industry-academia collaboration.
Conventional identification programs were unable to distinguish subtle differences in texture, so a new program had to be built. However, over the course of two years, an "automatic texture identification program" was developed that makes it possible to identify product numbers even for materials whose only distinguishing feature is minute differences in the surface.
In addition, given the current situation in which the construction industry has a high proportion of elderly people and people are reluctant to engage in complex operations or acquire specialized knowledge, the industry is not widespread. In response, the company developed an app called "Kabepita" that allows anyone to easily identify wallpaper product numbers, so that work can be completed with easy-to-understand buttons and intuitive operations.
The "Kabepita" app was awarded the Grand Prix for the following reasons: 6) It attempted to solve problems using image recognition AI through industry-academia collaboration; XNUMX) With the understanding and cooperation of major material manufacturers and six competing companies, it combined the company's accumulated expertise with Doshisha University's advanced academic technology to develop an "automatic texture identification program" that uses subtle differences in material surfaces as identification elements, successfully identifying the product numbers of popular wallpapers and turning them into an app; and XNUMX) In solving the identification problem and turning it into an app, Komatsu Corporation established a new DX Solutions Division and put in place a system for seamless collaboration with Doshisha University and partner companies.
