IMAGE INSPECTION METHOD, IMAGE INSPECTION DEVICE, AND IMAGE INSPECTION PROGRAM

To provide an image inspection method, an image inspection device, and an image inspection program capable of appropriately inspecting a state of a photographed image.SOLUTION: An image inspection method comprises steps of: acquiring S101 a photographed image CI; extracting S104 a plurality of featu...

Full description

Saved in:
Bibliographic Details
Main Authors KUHARA CHIKAKO, TAKAHASHI YOJI, CHIKATSU HIROBUMI
Format Patent
LanguageEnglish
Japanese
Published 22.04.2021
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:To provide an image inspection method, an image inspection device, and an image inspection program capable of appropriately inspecting a state of a photographed image.SOLUTION: An image inspection method comprises steps of: acquiring S101 a photographed image CI; extracting S104 a plurality of feature points P in the photographed image CI; calculating S105, for each of the plurality of feature points P, a gradient feature amount including a gradient direction average value, which is an average value in a gradient direction; generating S106 a histogram H showing distribution of number of occurrences of the average value in the gradient direction; calculating S107 a first kurtosis, which is a kurtosis of the histogram H in a first range of average values in the gradient direction, and a second kurtosis, which is a kurtosis of the histogram H in a second range of average values in the gradient direction different from the first range; and calculating S108 a quantitative value based on a first evaluation value, which is an absolute value of a difference between a sum of the first kurtosis and the second kurtosis and an absolute value of a difference between the first kurtosis and the second kurtosis.SELECTED DRAWING: Figure 3 【課題】撮像画像の状態を適切に検査することのできる画像検査方法、画像検査装置、及び画像検査プログラムを提供する。【解決手段】画像検査方法は、撮像画像CIを取得するステップS101と、撮像画像CIにおいて複数の特徴点Pを抽出するステップS104と、複数の特徴点Pのそれぞれについて、勾配方向の平均値である勾配方向平均値を含む勾配特徴量を算出するステップS105と、勾配方向平均値の出現数の分布を示すヒストグラムHを生成するステップS106と、勾配方向平均値の第1範囲におけるヒストグラムHの尖度である第1尖度と、第1範囲と異なる勾配方向平均値の第2範囲におけるヒストグラムHの尖度である第2尖度と、を算出するステップS107と、第1尖度と第2尖度との和と、第1尖度と第2尖度との差の絶対値と、の差の絶対値である第1評価値に基づいて定量値を算出するステップS108と、を備える。【選択図】図3
Bibliography:Application Number: JP20190188777