Show original
Try the app
Enjoyed this article?
Use "Request tipping" to ask the author to set up tip receiving.
Share
麻雀AIカメラを作った理由
AI translation
～Image Recognition Algorithm YOLO Application～
About 1 min read159
AI translation
～Image Recognition Algorithm YOLO Application～
Try the app
Enjoyed this article?
Use "Request tipping" to ask the author to set up tip receiving.
Mahjong point calculation, especially fu (符) calculation, is truly difficult for beginners. While learning it could be considered part of the fun, I wanted to make it possible for four people who can't calculate points to still enjoy mahjong.
Around the time COVID-19 lockdowns created free time, I started learning Python, and I saw videos of the YOLO algorithm performing real-time analysis of sports match videos (soccer), which I thought was amazing.
Not just mahjong tile image generation and hand pattern judgment, but also making it easy to set mahjong situations like prevailing wind, seat wind, riichi, ippatsu, rinshan, and haitei, so that points can be calculated simply. Additionally, for hands that are difficult for beginners like chinitsu, I made sure information is displayed showing what tiles need to be drawn from one-away-from-winning (iishantten) to form what kind of wait.
麻雀の点数計算、特に符計算は初心者には本当に難しい。それを覚えるのも楽しみの一つと言えなくもないが、点数計算ができない4人でも麻雀を楽しめるようにしたい
ちょうど、コロナ禍で暇ができたときにPythonの勉強を始めたところ、DeepLearningという技術が流行っていて、YOLOアルゴリズムでスポーツの試合（サッカー）動画をリアルタイム解析している動画を見てすげぇと思ったから
麻雀の牌画像生成と役判定だけではなく、場風、自風、リーチ・一発・嶺上・海底など麻雀の状況を簡単に設定して得点が計算できるようにしたことと、初心者には難しい清一色などで、イーシャンテンから何を引いたら、何待ちになるかなどの情報も表示されるようにした。
