This research addresses a constant problem in Poomsae competitions: the significant variation in accuracy ratings given by judges. We present the development of software based on artificial vision as a support tool for judges and competitors. The software integrates computer vision frameworks and specialized scoring algorithms to evaluate the accuracy of the Poomsae Pyongwon. In collaboration with the College of Referees of the State of Veracruz, we conducted a comparative evaluation of athletes across the Mexican Republic. In this evaluation, we contrast the ratings assigned by human judges with those provided by the software. The results indicated that, in most cases, the software's ratings were close to the mean of the ratings given by the human judges gave, demonstrating that the software can provide ratings close to expert judgment. Additionally, the software tended to be stricter due to the precise mathematical calculations of angles and distances. The software allows you to standardize qualifications and support training, benefiting judges and competitors. Likewise, it contributes to the scientific community by providing public data sets with information relevant to Taekwondo.