IP Library Granted Patent US 10,726,285
Granted Patent B2
US 10,726,285 · App. 16/165,111 · Granted Jul 28, 2020

Medicine audit apparatus, method, and program

Inventor: Kazuchika Iwami (Ashigarakami-gun, JP)
Assignee: FUJIFILM Toyama Chemical Co., Ltd.
G06K9/036A61J3/00A61J7/02G01N21/85G01N21/9508G06K9/00G06T7/0002G06T7/0012G06T7/13G16H20/13H04N5/23229A61J7/0454G06K9/2036G06K2209/01G06K2209/19G06T2207/20056G06T2207/30168H04N5/232
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,726,285
App. No.
16/165,111
Granted
Jul 28, 2020
Kind
B2
Abstract

Provided are a medicine audit apparatus, a medicine audit method, and a program capable of detecting a state where recognition of a medicine is impossible from a captured image obtained by imaging the medicine and urging a user to pay attention in the case of the state where the medicine recognition is impossible. The medicine audit apparatus 10 comprises a medicine recognition unit 54 that specifies a type of the medicine by recognizing the medicine from the captured image obtained by imaging the medicine, a quantitative value calculation unit 50 that calculates a quantitative value for quantifying a quality of the captured image related to success or failure of the medicine recognition by the medicine recognition unit 54 , a comparison determination unit 51 that compares the quantitative value calculated by the quantitative value calculation unit 50 with the threshold value to determine whether the recognition of the medicine is possible, and a warning output unit 58 that issues a warning in a case where the medicine recognition is determined to be impossible by the comparison determination unit 51.

Claims (51)

1. A medicine audit apparatus comprising:

an image acquisition unit that acquires a captured image of a medicine;

a medicine recognition unit that specifies a type of the medicine by recognizing the medicine from the captured image;

a quantitative value calculation unit that calculates a quantitative value for quantifying a quality of the captured image;

a storage unit that stores a threshold value;

a comparison determination unit that compares the quantitative value calculated by the quantitative value calculation unit with the threshold value to determine whether the recognition of the medicine by the medicine recognition unit is possible; and

a warning output unit that issues a warning in a case where the medicine recognition of the medicine recognition unit is determined to be impossible by the comparison determination unit,

wherein the medicine recognition unit includes:

an edge detection unit that detects an edge of the medicine from the captured image; and

an engraved mark recognition unit that recognizes an engraved mark assigned on the medicine extracted based on information on the edge detected by the edge detection unit from the captured image,

wherein the quantitative value quantifies the quality of the captured image related to edges of the captured image, and

wherein the threshold value is a determination criterion whether the edge detection by the edge detection unit is possible,

wherein the determination whether the recognition of the medicine by the medicine recognition unit is possible, which is performed by the comparison determination unit, is a determination whether the edge detection by the edge detection unit is possible, and

wherein the quantitative value calculation unit calculates the quantitative value using a ratio between a sum of absolute values of spatial frequency spectra of higher frequencies than one cycle/mm and a sum of absolute values of spatial frequency spectra of lower frequencies than one cycle/mm in the spatial frequency spectra obtained by performing processing of two-dimensional Fourier transform on the captured image in a radial direction.

2. The medicine audit apparatus according to claim 1 ,

wherein the quality of the captured image includes sharpness of the captured image.

3. The medicine audit apparatus according to claim 1 ,

wherein in a case where the quantitative value calculated by the quantitative value calculation unit is less than the threshold value, the warning output unit outputs the warning.

4. The medicine audit apparatus according to claim 1 ,

wherein the warning output unit includes a display unit that displays information on the warning.

5. The medicine audit apparatus according to claim 1 , further comprising:

a camera that images the medicine,

wherein the image acquisition unit acquires a captured image imaged by the camera.

6. The medicine audit apparatus according to claim 1 ,

wherein the captured image is an image obtained by imaging one-packaging medicines for one package in which a plurality of types of medicines are accommodated in a packaging bag.

7. A medicine audit method comprising:

an image acquisition step of acquiring a captured image of a medicine;

a medicine recognition step of specifying a type of the medicine by recognizing the medicine from the captured image;

a quantitative value calculation step of calculating a quantitative value for quantifying a quality of the captured image;

a comparison determination step of comparing a threshold value with the quantitative value calculated by the quantitative value calculation step to determine whether the recognition of the medicine by the medicine recognition step is possible; and

a warning output step of issuing a warning in a case where the medicine recognition of the medicine recognition step is determined to be impossible in the comparison determination step,

wherein the medicine recognition step includes:

an edge detection step of detecting an edge of the medicine from the captured image; and

an engraved mark recognition step of recognizing an engraved mark assigned on the medicine extracted based on information on the edge detected by the edge detection step from the captured image,

wherein the quantitative value is a quantitative value that quantifies the quality of the captured image related to edges of the captured image, and

wherein the threshold value is a determination criterion whether the edge detection by the edge detection step is possible,

wherein the determination whether the recognition of the medicine by the medicine recognition step is possible, which is performed by the comparison determination step, is a determination whether the edge detection by the edge detection step is possible, and

wherein the quantitative value calculation step calculates the quantitative value using a ratio between a sum of absolute values of spatial frequency spectra of higher frequencies than one cycle/mm and a sum of absolute values of spatial frequency spectra of lower frequencies than one cycle/mm in the spatial frequency spectra obtained by performing processing of two-dimensional Fourier transform on the captured image in a radial direction.

8. A recording medium that is a non-transitory computer-readable recording medium and records a program, in a case where the program stored in the recording medium is read by the computer, the program causing the computer to execute:

an image acquisition step of acquiring a captured image of a medicine;

a medicine recognition step of specifying a type of the medicine by recognizing the medicine from the captured image;

a quantitative value calculation step of calculating a quantitative value for quantifying a quality of the captured image;

a comparison determination step of comparing a threshold value with the quantitative value calculated by the quantitative value calculation step to determine whether the recognition of the medicine by the medicine recognition step is possible; and

a warning output step of issuing a warning in a case where the medicine recognition of the medicine recognition step is determined to be impossible in the comparison determination step,

wherein the medicine recognition step includes:

an edge detection step of detecting an edge of the medicine from the captured image; and

an engraved mark recognition step of recognizing an engraved mark assigned on the medicine extracted based on information on the edge detected by the edge detection step from the captured image,

wherein the quantitative value is a quantitative value for quantifying the quality of the captured image related to edges of the captured image, and

wherein the threshold value is a determination criterion whether the edge detection by the edge detection unit is possible,

wherein the determination whether the recognition of the medicine by the medicine recognition step is possible, which is performed by the comparison determination step, is a determination whether the edge detection by the edge detection step is possible, and

wherein the quantitative value calculation step calculates the quantitative value using a ratio between a sum of absolute values of spatial frequency spectra of higher frequencies than one cycle/mm and a sum of absolute values of spatial frequency spectra of lower frequencies than one cycle/mm in the spatial frequency spectra obtained by performing processing of two-dimensional Fourier transform on the captured image in a radial direction.

Assignments (3)
NUNC PRO TUNC ASSIGNMENT Recorded Sep 11, 2024
From: FUJIFILM TOYAMA CHEMICAL CO., LTD.
To: FUJIFILM MEDICAL CO., LTD.
Reel/Frame 068552/0594 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded May 20, 2019
From: FUJIFILM CORPORATION
To: FUJIFILM TOYAMA CHEMICAL CO., LTD.
Reel/Frame 049232/0326 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Oct 19, 2018
From: IWAMI, KAZUCHIKA
To: FUJIFILM CORPORATION
Reel/Frame 047237/0346 →
Priority Claims (1)
JP 2016-086032 · Apr 22, 2016 · national
Continuity (2)
Continuation PCTJP2017014261 · Apr 5, 2017
Related Publication 20190050660A1 · Feb 14, 2019