PROCESSING SYSTEM, PROCESSING METHOD AND PROGRAM

• Main Authors: KIKUCHI MASARU, NABETO YU, SHIRAISHI SOMA, SATO TAKAMI
• Format: Patent
• Language: English; Japanese
• Published: 25.03.2024
• Subjects: CALCULATING; CHECKING-DEVICES; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS; REGISTERING THE RECEIPT OF CASH, VALUABLES, OR TOKENS

To enhance the accuracy of image-based object recognition while reducing the processing burden on a computer that generates an estimation model.SOLUTION: A processing system according to the present invention includes: an image acquisition unit for acquiring a recognition processing image including an object; a recognition unit for recognizing the object in the recognition processing image based on an estimation model generated by machine learning; a correction reception unit for receiving an input for correcting the recognition result; a correction unit for storing, in a memory unit, correction information in which the corrected recognition result and the recognition processing image are associated with each other; and a learning unit for performing re-learning by using the recognition processing image stored as the correction information and updates an estimation model when the number of recognition processing images stored as the correction information exceeds a predetermined value.SELECTED DRAWING: Figure 4 【課題】推定モデルを生成するコンピュータの処理負担を軽減しつつ、画像に基づく物体認識の精度を高める。【解決手段】本発明は、物体を含む認識処理画像を取得する画像取得部と、機械学習で生成された推定モデルに基づき認識処理画像内の物体を認識する認識部と、認識の結果を訂正する入力を受付ける訂正受付部と、訂正後の認識の結果と認識処理画像とを紐付けた訂正情報を記憶部に記憶させる訂正部と、訂正情報として記憶された認識処理画像の数が所定値を超えると、訂正情報として記憶された認識処理画像を用いて再学習して推定モデルを更新する学習部と、を有する処理システムを提供する。【選択図】図４