• Patent Title: Method of detecting and quantifying blur in a digital image
  • Application No.: US17052107
    Application Date: 2019-04-16
  • Publication No.: US11321818B2
    Publication Date: 2022-05-03
  • Inventor: David AmeisenEmmanuel Ameisen
  • Applicant: IMGINIT
  • Applicant Address: FR Paris
  • Assignee: IMGINIT
  • Current Assignee: IMGINIT
  • Current Assignee Address: FR Paris
  • Agency: Klein, O'Neill & Singh, LLP
  • Priority: FR1853848 20180504
  • International Application: PCT/EP2019/059834 WO 20190416
  • International Announcement: WO2019/211094 WO 20191107
  • Main IPC: G06T5/00
  • IPC: G06T5/00 G06T7/00
Method of detecting and quantifying blur in a digital image
Abstract:
A method of detecting and quantifying blur in a digital image, making use of a computer and comprising: a step a) of obtaining a digital image comprising pixels; a step b) of obtaining a brightness parameter for each pixel, said step b) comprising a convolution operation with an edge detection matrix; a step c) of calculating a score S1 comprising the maximum, calculated over all the pixels, of the brightness parameter obtained in step b) and a second score S2 comprising a logistic function, and a step d) of evaluating the digital image, said digital image being considered to be blurry if the score S1 obtained in step c) is strictly less than a first predetermined threshold S10, the score S1 providing a first quantity of blur present in the digital image.
Public/Granted literature
Information query
Patent Agency Ranking
0/0