CN113658039A - Method for determining splicing sequence of label images of medicine bottles - Google Patents
Method for determining splicing sequence of label images of medicine bottles Download PDFInfo
- Publication number
- CN113658039A CN113658039A CN202110768956.6A CN202110768956A CN113658039A CN 113658039 A CN113658039 A CN 113658039A CN 202110768956 A CN202110768956 A CN 202110768956A CN 113658039 A CN113658039 A CN 113658039A
- Authority
- CN
- China
- Prior art keywords
- image
- label
- medicine bottle
- area
- medicine
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
- 239000003814 drug Substances 0.000 title claims abstract description 80
- 238000000034 method Methods 0.000 title claims abstract description 42
- 230000000877 morphologic effect Effects 0.000 claims abstract description 11
- 238000012545 processing Methods 0.000 claims abstract description 9
- 230000011218 segmentation Effects 0.000 claims abstract description 8
- 238000012937 correction Methods 0.000 claims description 16
- 238000003384 imaging method Methods 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 4
- 238000009434 installation Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 238000010561 standard procedure Methods 0.000 claims description 3
- 230000007547 defect Effects 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 5
- 238000004519 manufacturing process Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000005034 decoration Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 229940079593 drug Drugs 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 1
- 238000003702 image correction Methods 0.000 description 1
- 238000001802 infusion Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 230000037303 wrinkles Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/60—Rotation of whole images or parts thereof
- G06T3/608—Rotation of whole images or parts thereof by skew deformation, e.g. two-pass or three-pass rotation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
- G06T5/30—Erosion or dilatation, e.g. thinning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/155—Segmentation; Edge detection involving morphological operators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2200/00—Indexing scheme for image data processing or generation, in general
- G06T2200/32—Indexing scheme for image data processing or generation, in general involving image mosaicing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20092—Interactive image processing based on input by user
- G06T2207/20104—Interactive definition of region of interest [ROI]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
- Medical Preparation Storing Or Oral Administration Devices (AREA)
Abstract
The invention discloses a method for determining a splicing sequence of label images of a medicine bottle, and belongs to the technical field of image splicing. The splicing sequence determining method comprises the following steps: s1, acquiring an original multi-medicine bottle label image; s2, extracting a medicine bottle label image ROI area by utilizing threshold segmentation and morphological processing; s3, correcting the image; and S4, determining the splicing sequence. According to the invention, the splicing sequence of the camera images can be automatically determined by utilizing the template matching method and the priori knowledge, and the defects of the original fixed sequence splicing method are improved.
Description
Technical Field
The invention relates to a method for determining a splicing sequence of label images of a medicine bottle, and belongs to the technical field of image splicing.
Background
In recent years, many oral medicines and infusion medicines are packaged in medicine bottles. The bottle label contains important bottle information such as manufacturing company information, medicine name, bottle label number, and date of manufacture. Therefore, the existence or non-existence or missing of information contained in the label of the medicine bottle, the position of the label relative to the bottle body, the appearance of the label whether has damage, deformity, smearing, smudge or wrinkle and the like become important contents for detecting and identifying the medicine enterprises. Utilize machine vision automatic identification to detect medicine bottle label information, can not only reduce medical personnel's work load, raise the efficiency, also improved the accuracy greatly. Because the medicine bottle is cylindrical, a single common camera cannot completely show the corresponding view field in a wider view angle, and further cannot carry out 360-degree surrounding shooting; although a large visual angle can be obtained in long-distance shooting, the shot object is correspondingly shrunk, and the detection effect is easily influenced. Therefore, a high-definition panorama can be obtained at a low price, and panorama stitching of images is particularly important. At present, the splicing of cylindrical images is performed according to fixed sequence according to images shot by a camera, and the deviation of different angles of the medicine bottles relative to the imaging camera exists on a product line, so that the splicing sequence can be automatically determined by a splicing system on an automatic production line, and subsequent work such as label defect detection and character recognition is facilitated.
Disclosure of Invention
The invention aims to provide a method for determining a splicing sequence of label images of a medicine bottle, which aims to solve the problems in the prior art.
A splicing sequence determining method for label images of medicine bottles comprises the following steps:
s1, acquiring an original multi-medicine bottle label image;
s2, extracting a medicine bottle label image ROI area by utilizing threshold segmentation and morphological processing;
s3, correcting the image;
and S4, determining the splicing sequence.
Further, in S1, the method specifically includes the following steps:
s11, collecting pictures of the target object by adopting a plurality of cameras, placing the medicine bottles on an objective table, placing the cameras on four angles of a horizontal plane where the medicine bottles are located, enabling the lenses to be aligned with the medicine bottles, fixing the objective table, and manually rotating the medicine bottles by placing different medicine bottle samples to obtain imaging pictures at different angles;
and S12, removing the background interference part by using a fixed region cutting method according to the target motion condition and the imaging characteristics of the medicine bottle, thereby obtaining a coarse positioning subgraph.
Further, in S2, specifically, a segmentation method with a fixed threshold is used to obtain a binary image of the medicine bottle, the threshold T is 100, the label, the bottle cap, and the reflective region are completely separated, the region where the label is located in the binary result has the largest area, and the label target includes some holes, and then the region of the target is obtained by morphological processing to realize accurate extraction of the label, small-area interference regions are removed by area screening, 5 × 5 square structural elements are used to perform morphological closing operation, and the holes are refilled to obtain a result, and a highlight region is extracted to obtain an accurate label region.
Further, in S3, the method specifically includes the following steps:
s31, obtaining a medicine bottle image simultaneously according to the obtained minimum and maximum column information of the label area, wherein the medicine bottle is inclined at a certain angle in the vertical direction, the edge of the binary image is extracted in the vertical direction by using a Sobel operator, the rotation in the opposite direction is carried out after the angle is detected by using the Hough principle, and the bilinear interpolation operation is adopted in the rotation step for compensation reconstruction to obtain a corrected medicine bottle image and an accurate label area;
s32, performing cylindrical back projection correction on the target after inclination correction, wherein the calculation formula is shown as formula (1):
wherein, (x, y) is any pixel point on the back projection plane, W is the width of the image after back projection correction, (x ', y') is the corresponding cylindrical image pixel point, (W ', H') is the cylindrical image width and height respectively, and c is the camera focal length.
Further, in S4, specifically,
according to the installation condition of the camera, at most two images in 4 images obtained by shooting comprise a logo subgraph, wherein the logo on the right side of the image is the first image, the logo subgraph is taken as a template image to be matched, the four camera images are subjected to gray matching by adopting a normalized cross-correlation matching method NCC, and the calculation process is as follows:
assuming that the size of the target image S is m × m, the size of the template T is n × n, and m > n is satisfied, the template T is translated rightward in the target image S starting from the upper left corner, the difference between the image at each position and the template image is compared, and the difference between the target area and the template image is measured by a metric function, the metric function formula is shown in formula (2):
wherein, (x, y) is an image pixel point, f is a gray value of a pixel P, μ is an average value of all pixels of a window, a sigma index standard method, t is a pixel value of a template, an NCC value of 1 indicates that the correlation is high, a value of-1 indicates no correlation, pixels of a logo on a medicine bottle label are obtained according to the working principle of NCC, the mean value and the standard deviation are calculated, the window is moved from left to right on the image on the medicine bottle label image from top to bottom according to the size of the logo, the NCC values of the pixels in the window and the template pixels after each pixel is moved are calculated, a threshold th is set to be 0.55, if the NCC is larger than th, the matching is judged, the position column coordinate of the window at the moment is recorded, when the column coordinate is larger than half of the image width, the window is judged to be positioned at the right side of the whole target image, the first image to be spliced is obtained, and then according to the sequence of a camera, and obtaining a second graph, a third graph and a fourth graph to be spliced.
The invention has the following beneficial effects:
(1) the invention can be used for automatically determining the splicing sequence of a label of a medicine bottle or other bottle-shaped images, wherein the first image to be spliced is determined according to whether a logo appears on the label and the position of the logo appearing on the whole image by taking the logo image as a template image, and the splicing sequence can be automatically determined.
(2) The invention can also provide preprocessing means such as cylindrical back projection correction and inclination correction of the label.
Drawings
Fig. 1 is a flowchart of an implementation of a method for determining a splicing sequence of label images of a medicine bottle according to the present invention;
FIG. 2 is a simplified diagram of a label image capture system;
fig. 3 is a diagram showing the result of extracting a ROI region of a label of a medicine bottle, wherein fig. 3(a) is an original image; FIG. 3(b) is a coarse positioning diagram; FIG. 3(c) is a binary map of the tag region;
fig. 4 is a diagram showing a correction result of a label image of a medicine bottle, wherein fig. 4(a) is a medicine bottle image; fig. 4(b) is an image of the vial after tilt correction; fig. 4(c) is a precise label area diagram.
Here, 101 denotes a vial, 102 denotes a stage, 103 denotes a camera 4, 104 denotes a camera 3, 105 denotes a camera 2, and 106 denotes a camera 1.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, the present invention provides an embodiment of a method for determining a splicing sequence of label images of a medicine bottle, where the method for determining a splicing sequence includes the following steps:
s1, acquiring an original multi-medicine bottle label image;
s2, extracting a medicine bottle label image ROI area by utilizing threshold segmentation and morphological processing;
s3, correcting the image;
and S4, determining the splicing sequence.
Further, referring to fig. 2, in S1, the method specifically includes the following steps:
s11, collecting pictures of the target object by adopting a plurality of cameras, placing the medicine bottles on an objective table, placing the cameras on four angles of a horizontal plane where the medicine bottles are located, enabling the lenses to be aligned with the medicine bottles, fixing the objective table, and manually rotating the medicine bottles by placing different medicine bottle samples to obtain imaging pictures at different angles;
and S12, removing the background interference part by using a fixed region cutting method according to the target motion condition and the imaging characteristics of the medicine bottle, thereby obtaining a coarse positioning subgraph.
Specifically, a plurality of cameras are used for collecting pictures of a target object, the medicine bottle 101 is placed on the object stage 102, the cameras 103, 104, 105 and 106 are placed at four angles as shown in fig. 2, the object stage 102 is fixed, and different medicine bottle samples are placed, and the medicine bottle is manually rotated to obtain imaging pictures at different angles.
The bottle is located at the center of fig. 3(a), the bottle target occupies 1/3 size of the whole image, and the detailed information on the label is not beneficial to the subsequent feature extraction and other processes. Therefore, according to the target motion and the imaging characteristics of the medicine bottle, the background interference part is removed by using a fixed region clipping method, so as to obtain the rough positioning subgraph 3 (b).
Further, in S2, specifically, a segmentation method with a fixed threshold is used to obtain a binary image of the medicine bottle, in this embodiment, the threshold is selected as T100, the label, the bottle cap, and the reflective area are completely separated, the area of the area where the label is located in the binary result is the largest, meanwhile, the label target includes some holes, the area of the target is obtained by morphological processing to realize accurate extraction of the label, the small-area interference area is removed by area screening, the morphological closing operation is performed by using 5 × 5 square structural elements, the holes are refilled, the result is shown in fig. 3(c), and the highlighted area is extracted to obtain an accurate label area.
Further, in S3, the method specifically includes the following steps:
s31, obtaining a medicine bottle image according to the obtained minimum and maximum column information of the label region, as shown in fig. 4(a), at this time, the medicine bottle has a certain angle of inclination in the vertical direction, extracting the edge of the binary image in the vertical direction by using Sobel operator, detecting the angle by using hough principle, and then rotating in the opposite direction, wherein the rotation step uses bilinear interpolation operation to perform compensation reconstruction, so as to obtain a corrected medicine bottle image 4(b) and an accurate label region image 4 (c);
s32, the object after tilt correction has a certain cylindrical distortion, i.e. the object is convex outward in the middle, concave inward on both sides, and has a certain curvature on the bottom edge, as shown in fig. 4(c), and the image closer to the edge is less clear. Therefore, the cylindrical back projection correction needs to be performed on the target after the tilt correction, and the calculation formula is shown as formula (1):
wherein, (x, y) is any pixel point on the back projection plane, W is the width of the image after back projection correction, (x ', y') is the corresponding cylindrical image pixel point, (W ', H') is the cylindrical image width and height respectively, and c is the camera focal length.
Further, in S4, specifically,
due to the fact that the medicine bottle label rotates at any angle, the sequence of pictures shot each time cannot be determined, and the pictures need to be arranged in sequence in order to facilitate subsequent image processing. Given that the placement positions of the 4 cameras are fixed, only the first image to be spliced needs to be determined, and the sequence of the other three images can be determined according to the ranking of the cameras. According to the installation condition of the camera, at most two images in 4 images obtained by shooting comprise a logo subgraph, wherein the logo on the right side of the image is the first image, the logo subgraph is taken as a template image to be matched, the four camera images are subjected to gray matching by adopting a normalized cross-correlation matching method NCC, and the calculation process is as follows:
assuming that the size of the target image S is m × m, the size of the template T is n × n, and m > n is satisfied, the template T is translated rightward in the target image S starting from the upper left corner, the difference between the image at each position and the template image is compared, and the difference between the target area and the template image is measured by a metric function, the metric function formula is shown in formula (2):
wherein, (x, y) is an image pixel point, f is a gray value of a pixel P, μ is an average value of all pixels of a window, a sigma index standard method, t is a pixel value of a template, an NCC value of 1 indicates that the correlation is high, a value of-1 indicates no correlation, pixels of a logo on a medicine bottle label are obtained according to the working principle of NCC, the mean value and the standard deviation are calculated, the window is moved from left to right on the image on the medicine bottle label image from top to bottom according to the size of the logo, the NCC values of the pixels in the window and the template pixels after each pixel is moved are calculated, a threshold th is set to be 0.55, if the NCC is larger than th, the matching is judged, the position column coordinate of the window at the moment is recorded, when the column coordinate is larger than half of the image width, the window is judged to be positioned at the right side of the whole target image, the first image to be spliced is obtained, and then according to the sequence of a camera, and obtaining a second graph, a third graph and a fourth graph to be spliced.
The invention discloses a method for determining a splicing sequence of a label image of a medicine bottle, which can be applied to the pretreatment of the label image of the medicine bottle, and comprises the steps of obtaining four original medicine bottle images with different angles, and obtaining an ROI (region of interest) of the label of the medicine bottle by utilizing threshold segmentation and morphological processing; performing image correction, including inclination correction and deformation correction, on the label of the medicine bottle to obtain an image with the minimum difference from the original label image; and intercepting the Logo subgraph on the original label as a template image, acquiring the original line position information of the Logo subgraph, performing target matching on the four subgraphs by using a gray template matching method, recording the position of the camera image column with the highest score, and judging to obtain the first image to be spliced. According to the invention, the splicing sequence of the camera images can be automatically determined by utilizing the template matching method and the priori knowledge, and the defects of the original fixed sequence splicing method are improved.
The above embodiments are only used to help understanding the method of the present invention and the core idea thereof, and a person skilled in the art can also make several modifications and decorations on the specific embodiments and application scope according to the idea of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.
Claims (5)
1. A splicing sequence determining method for label images of medicine bottles is characterized by comprising the following steps:
s1, acquiring an original multi-medicine bottle label image;
s2, extracting a medicine bottle label image ROI area by utilizing threshold segmentation and morphological processing;
s3, correcting the image;
and S4, determining the splicing sequence.
2. The method for determining the splicing order of the label images of the medicine bottles according to claim 1, wherein in S1, the method specifically comprises the following steps:
s11, collecting pictures of the target object by adopting a plurality of cameras, placing the medicine bottles on an objective table, placing the cameras on four angles of a horizontal plane where the medicine bottles are located, enabling the lenses to be aligned with the medicine bottles, fixing the objective table, and manually rotating the medicine bottles to obtain imaging pictures at different angles by placing different medicine bottle samples;
and S12, removing the background interference part by using a fixed region cutting method according to the target motion condition and the imaging characteristics of the medicine bottle, thereby obtaining a coarse positioning subgraph.
3. The method for determining the splicing sequence of the label images of the medicine bottles according to claim 1, wherein in S2, specifically, a segmentation method with a fixed threshold is used to obtain a binary image of the medicine bottle, the threshold is selected as T100, the label, the bottle cap and the reflective area are completely separated, the area of the area where the label is located in the binary result is the largest, meanwhile, the label target includes some holes, then the area of the target is obtained by morphological processing, so as to realize accurate extraction of the label, a small-area interference area is removed by area screening, a 5 × 5 square structural element is used to perform morphological closing operation, the holes are refilled, a result is obtained, and a highlight area is extracted to obtain an accurate label area.
4. The method for determining the splicing order of the label images of the medicine bottles as claimed in claim 3, wherein in S3, the method specifically comprises the following steps:
s31, obtaining a medicine bottle image simultaneously according to the obtained minimum and maximum column information of the label area, wherein the medicine bottle is inclined at a certain angle in the vertical direction, the edge of the binary image is extracted in the vertical direction by using a Sobel operator, the rotation in the opposite direction is carried out after the angle is detected by using the Hough principle, and the bilinear interpolation operation is adopted in the rotation step for compensation reconstruction to obtain a corrected medicine bottle image and an accurate label area;
s32, performing cylindrical back projection correction on the target after inclination correction, wherein the calculation formula is shown as formula (1):
wherein, (x, y) is any pixel point on the back projection plane, W is the width of the image after back projection correction, (x ', y') is the corresponding cylindrical image pixel point, (W ', H') is the cylindrical image width and height respectively, and c is the camera focal length.
5. The method of determining a splicing order of label images for medicine bottles of claim 1, wherein in S4, specifically,
according to the installation condition of the camera, at most two images in 4 images obtained by shooting comprise a logo subgraph, wherein the logo on the right side of the image is the first image, the logo subgraph is taken as a template image to be matched, the four camera images are subjected to gray matching by adopting a normalized cross-correlation matching method NCC, and the calculation process is as follows:
assuming that the size of the target image S is m × m, the size of the template T is n × n, and m > n is satisfied, the template T is translated rightward in the target image S starting from the upper left corner, the difference between the image at each position and the template image is compared, and the difference between the target area and the template image is measured by a metric function, the metric function formula is shown in formula (2):
wherein, (x, y) is an image pixel point, f is a gray value of a pixel P, μ is an average value of all pixels of a window, a sigma index standard method, t is a pixel value of a template, an NCC value of 1 indicates that the correlation is high, a value of-1 indicates no correlation, pixels of a logo on a medicine bottle label are obtained according to the working principle of NCC, the mean value and the standard deviation are calculated, the window is moved from left to right on the image on the medicine bottle label image from top to bottom according to the size of the logo, the NCC values of the pixels in the window and the template pixels after each pixel is moved are calculated, a threshold th is set to be 0.55, if the NCC is larger than th, the matching is judged, the position column coordinate of the window at the moment is recorded, when the column coordinate is larger than half of the image width, the window is judged to be positioned at the right side of the whole target image, the first image to be spliced is obtained, and then according to the sequence of a camera, and obtaining a second graph, a third graph and a fourth graph to be spliced.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110768956.6A CN113658039A (en) | 2021-07-07 | 2021-07-07 | Method for determining splicing sequence of label images of medicine bottles |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110768956.6A CN113658039A (en) | 2021-07-07 | 2021-07-07 | Method for determining splicing sequence of label images of medicine bottles |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113658039A true CN113658039A (en) | 2021-11-16 |
Family
ID=78489162
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110768956.6A Withdrawn CN113658039A (en) | 2021-07-07 | 2021-07-07 | Method for determining splicing sequence of label images of medicine bottles |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113658039A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114627055A (en) * | 2022-02-14 | 2022-06-14 | 甘肃旭康材料科技有限公司 | Cylindrical label quality detection method and system |
CN117911668A (en) * | 2024-03-15 | 2024-04-19 | 深圳市力生视觉智能科技有限公司 | Drug information identification method and device |
-
2021
- 2021-07-07 CN CN202110768956.6A patent/CN113658039A/en not_active Withdrawn
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114627055A (en) * | 2022-02-14 | 2022-06-14 | 甘肃旭康材料科技有限公司 | Cylindrical label quality detection method and system |
CN117911668A (en) * | 2024-03-15 | 2024-04-19 | 深圳市力生视觉智能科技有限公司 | Drug information identification method and device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111474184B (en) | AOI character defect detection method and device based on industrial machine vision | |
CN108920992B (en) | Deep learning-based medicine label bar code positioning and identifying method | |
US8744173B2 (en) | Using extracted image text | |
Myers et al. | Rectification and recognition of text in 3-d scenes | |
CN105957015A (en) | Thread bucket interior wall image 360 DEG panorama mosaicing method and system | |
CN109961399B (en) | Optimal suture line searching method based on image distance transformation | |
CN111612765B (en) | Method for identifying and positioning round transparent lens | |
US20080226171A1 (en) | Correcting device and method for perspective transformed document images | |
US20020051575A1 (en) | Method and apparatus for recognizing text in an image sequence of scene imagery | |
JP6344389B2 (en) | Optical character recognition device | |
CN113658039A (en) | Method for determining splicing sequence of label images of medicine bottles | |
EP3051461B1 (en) | A computer implemented system and method for extracting and recognizing alphanumeric characters from traffic signs | |
Sun et al. | A visual attention based approach to text extraction | |
CN108961262B (en) | Bar code positioning method in complex scene | |
CN113903024A (en) | Handwritten bill numerical value information identification method, system, medium and device | |
CN111161295A (en) | Background stripping method for dish image | |
US9094617B2 (en) | Methods and systems for real-time image-capture feedback | |
CN110502948B (en) | Restoration method and device for folding two-dimensional code image and code scanning equipment | |
CN109993046B (en) | Self-shadow object edge identification method and device based on visual camera and vehicle | |
CN113505629A (en) | Intelligent storage article recognition device based on light weight network | |
Lijing et al. | A warped document image mosaicing method based on registration and TRS transform | |
KR101396713B1 (en) | Apparatus for recognizing medicine using image technique | |
CN110070490B (en) | Image stitching method and device | |
CN110610163B (en) | Table extraction method and system based on ellipse fitting in natural scene | |
JP2011107878A (en) | Position detection apparatus and position detection method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WW01 | Invention patent application withdrawn after publication | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20211116 |