Method and associated verification apparatus for verifying correct application of an image to a cylindrical container - Patents.com

Abstract

To provide a method of confirming accurate imparting of an image to a cylindrical container.SOLUTION: In a method of confirming accurate imparting of an image to a cylindrical container 135, processing means includes, via image acquisition means 210, an acquisition step of acquiring the image imparted to the cylindrical container 135, identification step of identifying at least one important marker contained in at least one image portion of the image, and verification step of verifying whether the identification of at least one important marker contained in at least one image portion of the image is successful. The processing means further includes a confirmation step of determining a region value corresponding to a region of at least one important marker in the image portion. When the region value is not included between a first value and a second value, the processing means generates signaling information, and the first value is different from the second value.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a method for checking the correct application of an image to a cylindrical container according to the preamble of claim 1. In particular, a method and an apparatus are disclosed herein for checking the correct application of an image, e.g. including texts and / or logos, to a substantially cylindrical container, whether directly reproduced or applied, e.g. by means of an adhesive film. Such containers may be, e.g., vials, phials, bottles, cans, etc., adapted to contain, e.g., liquids or powders, such as medicines, perfumes, oils, etc. The present invention may be advantageously used in plants for producing, e.g., glass vials, adapted to contain medicines, for automatically identifying and / or discarding vials with incorrectly applied texts and / or logos. [Background technology]

[0002] A paper by Tao Ni et al., published in Advances in Intelligent Systems Research, volume 133 (AIIE 2016), entitled "Cylindrical Medicine Bottle's Character Recognition of Automotive Dispensing Machine", describes a system for recognizing labels on medicines contained in small cylindrical bottles. Such a system can rectify an image of such a label captured by a video camera from a cylindrical surface to a flat surface. Such rectified images are then used to recognize characters via techniques such as OCR (Optical Character Recognition) and using SIFT (Scale Invariant Feature Transform) descriptors, which are well known in the art.

[0003] The above mentioned label recognition systems known in the art suffer from several drawbacks as set out below.

[0004] A first drawback is that an image, e.g. including text and / or logos, applied to such a substantially cylindrical container may have erroneously blurred or missing parts, e.g. due to the process of printing such text and / or logos on the container. The technique proposed in the above paper allows for the identification of such text, e.g. letters and / or ideograms and / or logos, but does not allow for the detection of blurred or missing parts in the text and / or logos, and therefore a user may find it difficult to understand such text and / or logos.

[0005] Another drawback is that an image, including for example texts and / or logos, may be reproduced on the adhesive film, but then the image may be applied to the substantially cylindrical container inappropriately, for example with bulges and / or parts protruding from the cylindrical container, which may result in the texts and / or logos applied to the cylindrical container being altered, even if only partially. The technique proposed in the above paper allows the identification of such texts, for example letters and / or ideograms and / or logos, but does not allow the detection of the presence of bulges and / or protrusions in the texts and / or logos applied to the cylindrical container, so that a user may find it difficult to understand such texts and / or logos.

[0006] A further disadvantage is that such blurred or missing parts, or such bulges and / or protrusions, are particularly important for the accurate understanding of images that contain, for example, ideographic characters. Summary of the Invention [Problem to be solved by the invention]

[0007] It is therefore an object of the present invention to solve these and other problems of the prior art, in particular to provide a method and device for verifying the correct application of an image to a substantially cylindrical container, which makes it possible to detect the presence of any defects, such as, for example, blurred or missing parts, in the text and / or logo of the image printed on the container.

[0008] Another object of the invention is to provide a method and device for checking the correct application of an image to a substantially cylindrical container, which makes it possible to detect any defects resulting from the presence of bulges and / or protrusions in texts and / or logos applied to the cylindrical container, for example by an adhesive film.

[0009] It is a further object of the present invention to provide a method and apparatus for verifying the correct application of images to substantially cylindrical containers, which allows for efficient real-time detection of the presence of such defects. [Means for solving the problem]

[0010] The invention described herein comprises a method and apparatus for verifying accurate application of images to substantially cylindrical containers in real time, in particular by analyzing images of such cylindrical containers acquired during the manufacturing process.

[0011] Further advantageous features of the invention are set out in the appended claims, which form an integral part of this specification.

[0012] The present invention will now be described in detail through some non-limiting exemplary embodiments thereof, with particular reference to the accompanying drawings. [Brief description of the drawings]

[0013] [Figure 1]FIG. 2 is a schematic diagram illustrating an example of a system for processing substantially cylindrical containers, comprising a verification device for verifying correct application of an image to such containers, in accordance with an embodiment of the present invention. [Diagram 2] 2 is an exemplary block diagram of the verification device of FIG. 1; [Diagram 3] 3 illustrates an exemplary flow chart of a method for verifying correct application of an image to a cylindrical container in association with the verification device of FIG. 2. [Figure 4a] FIG. 1 shows an example of an image to be applied to a cylindrical container. [Figure 4b] FIG. 13 shows an example of a portion of an image to be applied to a cylindrical container. [Figure 4c] FIG. 13 shows an example of an ideogram that may be included in an image to be applied to a cylindrical container. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] 1, which shows diagrammatically a system 100 for processing one or more substantially cylindrical containers 135, such as vials, phials, bottles, and cans, made of, for example, glass, plastic, or other suitable material. Said system 100 comprises, for example, an image applying unit 160, a first storage unit 171, a second storage unit 172, a conveyor means 150, and at least one verification device 200 for verifying correct application of an image to such cylindrical containers 135.

[0015] The first storage unit 171 is adapted to store at least one cylindrical container 135, e.g. a medicinal glass vial, before an image 410, shown in Fig. 4a, containing information about such medicine is applied thereto. The image 410 may include at least one important sign S, e.g. at least one alphanumeric character, at least one ideographic character, at least one logo, etc. The first storage unit 171 may comprise actuator means for moving the at least one cylindrical container 135 outwards, which may comprise an electric motor and / or a servo mechanism driven by a hydraulic system.

[0016] The image-applying unit 160 is adapted to apply an image 410 to at least one cylindrical container 135, for example to an outer surface of said cylindrical container 135. The image-applying unit 160 may comprise actuator means adapted to perform a printing process on the at least one cylindrical container 135 and / or actuator means for applying an adhesive film comprising said image 410 to the at least one cylindrical container 135. The first storage unit 171 and said image-applying unit 160 may be operatively connected to each other such that at least one cylindrical container 135 can be moved from the first storage unit 171 to the image-applying unit 160. The image-applying unit 160 may comprise actuator means for moving out the at least one cylindrical container 135 with said image 410 applied thereto, which may comprise a servo mechanism driven by an electric motor and / or a hydraulic system.

[0017] The second storage unit 172 is adapted to store at least one cylindrical container 135, e.g. a pharmaceutical glass vial, having said image 410 applied to its outer surface. The second storage unit 172 may comprise actuator means for moving in the at least one cylindrical container 135 having said image 410 applied thereto, said actuator means comprising an electric motor and / or a servo mechanism driven by a hydraulic system.

[0018] The conveyor means 150 is adapted to transport at least one cylindrical container 135 having said image 410 applied thereto from the image application unit 160 to the second storage unit 172. The conveyor means 150 may be operatively connected to the image application unit 160 and the second storage unit 172 and may comprise, for example, a belt, a chain, etc., e.g. made of rubber and / or metal. The conveyor means 150 may comprise a servo mechanism driven by an electric motor and / or a hydraulic system.

[0019] The verification device 200 of the present invention is adapted to verify correct application of said image 410 to said at least one cylindrical container 135. The verification device 200 may, for example, be operatively connected to the conveyor means 150 for moving the at least one cylindrical container 135, for example, from the image application unit 160 to the second storage unit 172.

[0020] However, the image applying unit 160 may not be included in the system 100. This may be the case if the cylindrical container 135 contained in the first storage unit 171 has been previously applied with an image 410.

[0021] Figure 2 shows an exemplary block diagram of a verification device 200 associated with the system 100 of Figure 1. The verification device 200 may comprise an image acquisition means 210, an actuator means 220, an interface means 230, a memory means 240, and a processing means 250, which may be operatively interconnected via a communication bus 201.

[0022] The image acquisition means 210 is adapted to acquire said image 410 applied to said cylindrical container 135. The image acquisition means 210 may for example comprise a video camera, preferably a linear video camera, or a photo camera, preferably a high resolution photo camera, and a lamp, preferably an infrared lamp, adapted to illuminate the cylindrical container 135. In particular, the image acquisition means 210 may comprise a shallow depth of field optical system adapted to capture at least one portion of the image 410 located on a surface of the cylindrical container 135 exposed towards the image acquisition means 210. The image acquisition means 210 may transmit data of the image 410 to said memory means 240 and / or to said processing means 250. The data of the image 410 may be encoded according to a video coding standard, for example H264, H265, or an image coding standard, for example JPEG, or the like. Alternatively, the data of the image 410 may be in an uncompressed format, for example bitmap format, PNG format, or the like.

[0023] The actuator means 220 is adapted to move the cylindrical container 135 inside the verification device 200. The actuator means 220 may for example rotate the cylindrical container 135 around its axis of symmetry so that said image acquisition means 210 can easily acquire the image 410 applied to the cylindrical container 135. The actuator means 220 may comprise a servo mechanism driven by an electric motor and / or a hydraulic system, for example comprising a rubber coated wheel and / or a cylinder and an arm adapted to temporarily disengage the cylindrical container 135 from the conveyor means 150 to facilitate the acquisition of the image 410 applied to the cylindrical container 135.

[0024] The interface means 230 are adapted to output from the verification device 200 information obtained by the method for verifying the correct application of the image 410, which is the subject of the present invention, as described below with reference to the flow chart of FIG. 3. The interface means 230 may comprise, for example, a communication unit adapted to communicate with a remote management system and / or server. Said communication unit may comprise, for example, an Ethernet interface, a Wi-Fi interface, a GSM interface, etc. The communication unit may establish a connection with an external device for managing or monitoring the system 100, such as, for example, a computer, a smartphone, a tablet, etc. The interface means 230 may allow a user to interact with the verification device 200. For example, the interface means 230 may comprise output and input means, such as, for example, a display and an alphanumeric keyboard, respectively, or alternatively, an alphanumeric keyboard and a touch screen displaying interactive symbols. In another embodiment of the present invention, the interface means 230 may comprise a communication port for a terminal external to the verification device 200, such as, for example, an RS232 interface, a USB interface, etc. The external terminal of the verification device 200 may be, for example, a smartphone controlled by a user or an operator.

[0025] The memory means 240 allows the storage of information to be input and / or output to the verification device 200 and instructions implementing this embodiment of the invention. The memory means 240 may comprise, for example, a flash type solid-state memory. Said information may comprise, for example, values ​​and / or parameters, such as the operating state of the verification device 200 and / or said system 100, and / or a set of values ​​of a number of physical quantities, such as, for example, the speed of the cylindrical container 135 entering and / or leaving the verification device 200, the rotation speed of the cylindrical container 135, the intensity of light incident on the container 135, etc. The instructions stored in the memory means 240 are explained in more detail below with reference to the flow chart of FIG. 3.

[0026] The processing means 250 allows to process information and instructions stored in the memory means 240 and / or received from said image acquisition means 210, actuator means 220, interface means 230 and may comprise, for example, an ARM processor, an Arduino microcontroller, a processor with x86 architecture, etc.

[0027] An exemplary method for verifying correct application of image 410 to at least one cylindrical container 135 in association with verification device 200 of FIG. 2 is described below with reference to FIG. 3 and FIGS. 4a, 4b, and 4c.

[0028] At step 300, a step of receiving at least one cylindrical container 135 is performed, in which for example the processing means 250 controls the actuator means 220 to pick up the cylindrical container 135 from the conveyor means 150 and facilitate said image capture means 210 to capture an image 410 applied to the cylindrical container 135.

[0029] In step 310, the processing means 250 is configured to perform an image acquisition stage. During this stage, the processing means 250 acquires, via the image acquisition means 210, said image 410 applied to said cylindrical container 135, for example applied to the outer surface of said cylindrical container 135. During this stage, the processing means 250 can control, via the actuator means 220, within the working range of said image acquisition means 210, a movement of at least one cylindrical container 135, for example a rotation of the cylindrical container 135 around its axis of symmetry. During this stage, the processing means 250 may modify the acquired image 410, for example according to the teachings of the above-mentioned paper by Tao Ni et al. During this stage, the processing means 250 may segment the acquired image 410, for example according to a predefined segmentation scheme or alternatively according to an automatic segmentation scheme, to obtain at least one image portion 415. The segmentation of the image 410 is realized by the processing means 250 in such a way that at least one significant sign S, for example an alphanumeric character, an ideogram or a logo, is included in said image portion 415. During this stage, the image portions 420 may be stored, even if only partially, in the memory means 240 and subsequently processed by the processing means 250.

[0030] In step 320, the processing means 250 are configured to perform an identification step. During this step, the processing means 250 identify at least one important sign S contained in at least one image portion 415 of the image 410. The identification of the at least one important sign S may be achieved, for example, using techniques such as pattern matching, OCR, SIFT, etc. During this step, the image portion 415 is compared with at least one reference image of a set of reference images. Each reference image comprises at least one important sign S, to which an identifier is associated. Said set of reference images may be stored in the memory means 240, for example as a database in any compressed or uncompressed format. The result of the identification step may be, for example, a numerical value representative of the similarity of the important sign S contained in said image portion 415 with the important sign S of each reference image. Said numerical value may, for example, be comprised between "0" and "1", where "0" indicates no similarity and "1" indicates perfect similarity. The result of the identification step may be, for example, a list of identifiers sorted in descending order based on said numerical value, or alternatively, the identifier associated with the significance indicator S corresponding to the maximum value of said numerical value.

[0031] During this process, another embodiment of the invention may also comprise a counting step in which the processing means 250 determines the number of tracts of all tracts T of said at least one important sign S in the image portion 415. Said tracts are for example tracts constituting alphanumeric characters, ideograms, logos, etc. The ideogram shown in Fig. 4c illustrates an important sign S composed of three tracts T. If any imperfections are present, the number of tracts obtained from said counting step may be, for example, a number larger or smaller than three.

[0032] In step 330, the processing means 250 is configured to execute a verification step. During this step, the processing means 250 verifies whether the identification of at least one important sign S contained in at least one image portion 415 has been successful. For example, such verification may be performed by comparing the highest value of said numerical values ​​with a predefined threshold value Vs. If the highest value is above such threshold value Vs, it means that the recognition has been successful and the processing means 250 executes step 340. In case of failure of the identification of said at least one important sign S, i.e. when the highest value is below such threshold value Vs, the processing means 250 generates signaling information and executes step 355. The signaling information may for example comprise a binary or alphanumeric error code indicating that the identification of said at least one important sign S has not been successful.

[0033] During this step, in the above-mentioned other embodiment of the present invention, the processing means 250 may also be configured to verify whether said number of tracts corresponds to the number of tracts of said reference image associated with said important landmark S. For example, in addition to said identifier, a corresponding number of tracts of the reference image may be associated with each reference image. If said number of tracts corresponds to the number of tracts of the reference image identified during said identification phase, the processing means 250 executes step 340. Otherwise, if said number of tracts does not correspond to the number of tracts of the reference image associated with said important landmark S, the processing means 250 may generate signaling information comprising said number of tracts and an identifier corresponding to at least one identified important landmark S, and subsequently executes step 355.

[0034] In step 340, the processing means 250 is adapted to perform a verification step. During this step, the processing means 250 determine an area value A corresponding to the area of ​​said at least one important sign S in said image portion 415. For example, if the image portion 415 is an 8-bit coded greyscale image, each pixel of the image portion 415 may be coded as a value ranging from 0 to 255. In such a case, after setting a given pixel value threshold, for example 128, pixels corresponding to said important sign S may be identified, for example, using a value comprised between 0 and 128, while pixels corresponding to the background of the image portion 415 may be identified, for example, using a value comprised between 129 and 255. The processing means 250 may then determine said area value A by counting the pixels constituting said important sign S in said image portion 415, i.e. by counting the pixels comprised between 0 and 128. In another embodiment of the invention, said pixel value threshold may be predefined or determined on the basis of the acquired image 410, for example by calculating the average value of all pixel values ​​of the image 410.

[0035] In step 350, the processing means 250 generates a signaling information if said area value A determined in step 340 is not included between a first value V1 and a second value V2, where the first value V1 is different from the second value V2. The signaling information may comprise the area value A and an identifier corresponding to the at least one important sign S identified in step 330. The first value V1 and the second value V2 may be determined based on at least one reference image associated with said important sign S. Such values ​​V1 and V2 may for example be determined as a percentage of the area corresponding to the at least one important sign S of the reference image. For example, denoting the area in pixels corresponding to the at least one important sign S of the reference image as AR, such values ​​may be defined as V1=90%AR=0.90×AR and V2=110%AR=1.10×AR. Such percentage values ​​may depend on the type of image 410 applied to the cylindrical container 135, the reading process, etc. The method according to the invention is computationally less demanding and advantageously allows for example to detect in real time any defects, identified in terms of area, measured in pixels, in at least one significant landmark S reproduced in the image portion 415, and to generate signaling information if such a defect exceeds a tolerance range defined by values ​​V1 and V2. If said area value A is not included between the first value V1 and the second value V2, the processing means 250 executes step 355, otherwise it executes step 360.

[0036] During this step, in the above embodiment of the present invention, the signaling information may include the tract number of at least one significant marker S of the image portion 415 .

[0037] In step 355, the processing means 250 outputs said signaling information via said interface means 230. For example, the signaling information may be displayed on a screen so that an operator can monitor the correct application of the image 410 to the cylindrical container 135. Such signaling information may actuate the actuator means 220 to discard the cylindrical container 135 having at least one image portion 415 that was not identified or has a number of unacceptable defects. For example, the cylindrical container 135 may be discarded by one or more pneumatic cylinders included in the actuator means 220, and the discarded cylindrical container 135 may be placed in a collection chamber. Such signaling information may be transmitted to a management system of the system 100, even a remote management system, for example via a communication interface such as Wi-Fi, GSM, Ethernet, etc. The processing means 250 then executes step 360.

[0038] In step 360, the processing means 250 verifies whether all image portions 415 obtained during the acquisition stage described in step 310 have been processed according to the above steps 320, 330, 340, 350 and 355. If so, the processing means 250 performs step 370, if not, the processing means 250 performs step 320.

[0039] In step 370, the processing means 250 performs the step of releasing at least one cylindrical container 135. In this step, for example, the processing means 250 controls the actuator means 220 to place the cylindrical container 135 on the conveyor means 150 which transports the cylindrical container 135 towards the second storage unit 172.

[0040] The advantages of the present invention are apparent from the above description.

[0041] The method and associated verification device for verifying the correct application of an image to a substantially cylindrical container according to the present invention advantageously enables detection of the presence of any defects, such as, for example, blurred or missing portions, in the text and / or logo of the image applied to the container.

[0042] Another advantage of the present invention is that it is possible to detect the presence of any bulging and / or protruding portions of text and / or logos applied to a cylindrical container by an adhesive film.

[0043] A further advantage of the present invention is that it provides a method and associated verification apparatus for verifying the correct application of an image to a substantially cylindrical container that enables efficient, real-time detection of the presence of defects in the text and / or logo of the image.

[0044] Naturally, the embodiments and implementation details may be widely varied from those described and illustrated herein, purely by way of non-limiting examples, without violating the principles of the invention, but without departing from the scope of protection of the invention as set forth in the appended claims.

Claims

1. 1. A method for verifying accurate application of an image to a cylindrical container, the method comprising: an acquisition step in which a processing means acquires, via an image acquisition means, the image applied to the cylindrical container; an identification step in which said processing means identifies at least one significant feature contained in at least one image portion of said image; a verification step in which said processing means verifies whether the identification of at least one significant landmark contained in said at least one image portion has been successful; a counting step in which the processing means determines a tract number of all tracts of the at least one landmark of interest in the image portion; The method comprises a confirmation step in which the processing means determines an area value corresponding to an area of ​​the at least one important landmark in the image portion, the area value being a count value of pixels constituting the at least one important landmark, the processing means generating signaling information if the area value is not included between a first value and a second value, the first value being different from the second value, and if the number of tracts of the at least one important landmark in the image portion does not match the number of tracts of a reference image associated with the at least one important landmark, the processing means generating the signaling information including the number of tracts.

2. The method of claim 1 , wherein the first value and the second value are determined based on at least one reference image associated with the landmark of interest.

3. 3. A method according to claim 1 or 2, wherein the processing means determines the area value by counting pixels that constitute the landmark in the image portion.

4. The method according to any one of claims 1 to 3, wherein the signalling information comprises the domain value and an identifier corresponding to the at least one significant indicator.

5. A method according to any one of claims 1 to 4, wherein said processing means outputs said signalling information via an interface means.

6. 1. A verification device adapted to verify correct application of an image to a cylindrical container, said verification device comprising image acquisition means and processing means, acquiring the image provided on the cylindrical container via the image acquisition means; identifying at least one significant landmark contained in at least one image portion of the image; Verifying whether the at least one significant landmark included in the at least one image portion has been successfully identified; image acquisition means and processing means adapted to determine a tract number of all tracts of said at least one landmark of interest in said image portion; In the verification device, the processing means is adapted to determine an area value corresponding to an area of ​​the at least one important landmark in the image portion, the area value being a count value of pixels constituting the at least one important landmark, and the processing means is adapted to generate signaling information if the area value is not included between a first value and a second value, the first value being different from the second value, and if the number of tracts of the at least one important landmark in the image portion does not match the number of tracts of a reference image associated with the at least one important landmark, the processing means generates the signaling information including the number of tracts.

7. The verification device of claim 6 , wherein the first value and the second value are determined based on at least one reference image associated with the landmark of interest.

8. 8. An apparatus as claimed in claim 6 or 7, wherein the processing means is adapted to determine the area value by counting pixels constituting the landmark in the image portion.

9. Verification device according to any one of claims 6 to 8, wherein said signalling information comprises said domain value and an identifier corresponding to said at least one significant indicator.

10. An authentication device according to any one of claims 6 to 9, wherein the processing means is adapted to output the signalling information via an interface means.

11. The verification device according to any one of claims 6 to 10, wherein the image acquisition means comprises an optical system having a depth of field adapted to capture at least one image portion disposed on a surface of the cylindrical container, the at least one image portion being exposed towards the image acquisition means.

12. 1. A system for processing one or more substantially cylindrical containers, comprising: a conveyor means adapted to transport said cylindrical containers having an image applied thereto; and at least one verification device operatively connected to said conveyor means, said verification device comprising image acquisition means and processing means, acquiring the image provided on the cylindrical container via the image acquisition means; identifying at least one significant landmark contained in at least one image portion of the image; Verifying whether the at least one significant landmark included in the at least one image portion has been successfully identified; image acquisition means and processing means adapted to determine a tract number of all tracts of said at least one landmark of interest in said image portion; In the system, the processing means of the verification device is adapted to determine an area value corresponding to an area of ​​the at least one important landmark in the image portion, the area value being a count value of pixels constituting the at least one important landmark, and the processing means is adapted to generate signaling information if the area value is not included between a first value and a second value, the first value being different from the second value, and if the number of tracts of the at least one important landmark in the image portion does not match the number of tracts of a reference image associated with the at least one important landmark, the processing means generates the signaling information including the number of tracts.

13. The system of claim 12 , wherein the first value and the second value are determined based on at least one reference image associated with the landmark of interest.

14. 14. A system according to claim 12 or 13, wherein the processing means is adapted to determine the area value by counting pixels constituting the landmark in the image portion.

15. The system of any one of claims 12 to 14, wherein the signaling information comprises the domain value and an identifier corresponding to the at least one significant indicator.

16. A system according to any one of claims 12 to 15, wherein the processing means is adapted to output the signalling information via an interface means.

Images