Description:FIELD OF INVENTION
[0001] Embodiments of a present disclosure relate to a field of product inspection and more particularly to a system and a method for counting and detecting defects in the articles.
BACKGROUND
[0002] A product inspection refers to a systematic process of checking the quality of a product based on a specified set of standards. Product inspections allow you to confirm product quality on-site at different stages of the production process and before its shipment. Inspecting the products before it leaves the manufacturer's premises is an effective way of preventing quality problems and supply chain disruptions further down the line. Therefore, inspection for quality is an important and a critical aspect of the product or an article manufacturing industry.
[0003] Random manual inspection on samples of articles such as bottles moving on high-speed conveyor is a highly error prone process. Automatic inspection of bottles on high-speed moving conveyor is the pain point of the manufacturing industry. Also, counting bottles at each stage to understand where and why bottles were rejected helps to track and optimize inventory real time.
[0004] There is a need for a system that provides a real time insight about the cause of defects as and when the defects are detected to the manufacturers. Also, there is a need for a system which suggests a corrective action for the defect and that may be immediately taken to minimize rejections of the articles. Further, there is need for an automated system for detecting articles and defects therein.
[0005] Hence, there is a need for a system and a method for counting articles and detecting defects in the articles that addresses the aforementioned issues.
OBJECTIVE OF THE INVENTION
[0006] An objective of the present invention is to provide an automated system for counting articles and detecting defects in the articles.
[0007] Another objective of the present invention is to facilitate tracking and optimizing an inventory of the articles in real-time.
[0008] Yet, another objective of the present invention is to provide real-time insight into the cause and area of occurrence of the defect in the article.
[0009] Further, an objective of the present invention is to predict a corrective measure for the defective article.
BRIEF DESCRIPTION
[0010] In accordance with one embodiment of the disclosure, a system for counting articles and detecting defects in the articles based on machine learning is provided. The system includes a processing subsystem, hosted on a server, and configured to execute on a network to control bidirectional communications among a plurality of modules. The plurality of modules includes an article detection module, a counting module, a defect detection module, a prediction module, an identification module, and a recommendation module. The article detection module is configured to capture at least one of a video and an image of the plurality of articles moving on a conveyor belt in real-time. The article detection module is also configured to process the at least one of video and image for noise removal, background subtraction and resize of the image. Further, the article detection module is configured to provide a bounding box across a region of interest of the article to detect the article. The counting module is operatively coupled with the article detection module. The counting module is configured to count one or more regions of interest to count the number of articles moving on the conveyor belt. The defect detection module is operatively coupled with the article detection module. The defect detection module includes a visual unit for generating three-dimensional images of the article. The generated three-dimensional images are used to detect one or more defects in the article. The detected defects are stored in a repository. The prediction module is operatively coupled with the defect detection module. The prediction module is configured to predict one or more corrective measures for the defective article by comparing the detected defects with the stored defects in the repository. The prediction is based on a machine learning model collect the real-time data of the detected defects with respect to the predicted corrective measures. The prediction module is configured to send the collected data to an article making machine for enabling active learning. The article making machine is operatively coupled with the processing subsystem. The predicted corrective measures are stored in a database. The identification is module operatively coupled with the prediction module, wherein the identification module is configured to identify an anomaly of the article making machine. The identification of the anomaly is based on internet of things sensors connected to the article making machine. The recommendation module operatively coupled with the prediction module and the identification module. The recommendation module is configured to recommend a plurality of preventive measures by comparing with the predicted corrective measures stored in the database. The plurality of preventive measures is recommended in a plurality of languages. The recommendation module is also configured to provide feedback on the recommended plurality of corrective measures to the article making machine.
[0011] In accordance with another embodiment, a method for counting articles and detecting defects in an article is provided. The method includes capturing, by an article detection module of a processing subsystem, capture at least one of a video and an image of the plurality of articles moving on a conveyor belt in real-time. The method also includes processing, by the article detection module of the processing subsystem, the at least one of video and image for noise removal, background subtraction and resize of the image. Further, the method includes providing, by the article detection module of the processing subsystem, a bounding box across a region of interest of the article to detect the article. Furthermore, the method includes counting, by a counting module of the processing subsystem, one or more regions of interest to count the number of articles moving on the conveyor belt. Furthermore, the method includes generating, by a visual unit of a defect detection module of the processing subsystem, three-dimensional images of the article. Furthermore, the method includes detecting, by the defect detection module of the processing subsystem, one or more defects in the article. Furthermore, the method includes storing by a repository of the defect detection module of the processing subsystem, the detected defects. Furthermore, the method includes predicting, by a prediction module of the processing subsystem, one or more corrective measures for the defective article by comparing the detected defects with the stored defects in the repository, wherein the predicting is based on a machine learning model. Furthermore, the method includes collecting, by the prediction module of the processing subsystem, the real-time data of the detected defects with respect to the predicted corrective measures. Furthermore, the method includes sending, by the prediction module of the processing subsystem, the collected data to an article making machine for enabling active learning, wherein the article making machine is operatively coupled with the processing subsystem. Furthermore, the method includes storing, by the prediction module of the processing subsystem, the predicted corrective measures in a database. Furthermore, the method includes identifying, by an identification module of the processing subsystem, an anomaly of the article making machine, and wherein the identification of the anomaly is based on internet of things sensors connected to the article making machine. Furthermore, the method includes recommending, by a recommendation module of the processing subsystem, a plurality of preventive measures by comparing with the predicted corrective measures stored in the database, wherein the plurality of preventive measures is recommended in a plurality of languages. Furthermore, the method includes providing, by the recommendation module of the processing subsystem, feedback on the recommended plurality of corrective measures to the article making machine.
[0012] To further clarify the advantages and features of the present disclosure, a more particular description of the disclosure will follow by reference to specific embodiments thereof, which are illustrated in the appended figures. It is to be appreciated that these figures depict only typical embodiments of the disclosure and are therefore not to be considered limiting in scope. The disclosure will be described and explained with additional specificity and detail with the appended figures.
BRIEF DESCRIPTION OF THE DRAWINGS
[0013] The disclosure will be described and explained with additional specificity and detail with the accompanying figures in which:
[0014] FIG. 1 is a block diagram representing a system for counting articles and detecting defects in articles in accordance with an embodiment of the present disclosure;
[0015] FIG. 2 is a block diagram representing an exemplary embodiment the system for counting articles and detecting defects in articles with an embodiment of the present disclosure;
[0016] FIG. 3 is a block diagram of a computer or a server for of the system for counting articles and detecting defects in articles in accordance with an embodiment of the present disclosure;
[0017] FIG. 4a is a flow chart representing steps involved in a method for counting articles and detecting defects in articles in accordance with an embodiment of the present disclosure; and
[0018] FIG. 4b is a illustrates continued steps of the method of FIG. 4a in accordance with an embodiment of the present disclosure.
[0019] Further, those skilled in the art will appreciate that elements in the figures are illustrated for simplicity and may not have necessarily been drawn to scale. Furthermore, in terms of the construction of the device, one or more components of the device may have been represented in the figures by conventional symbols, and the figures may show only those specific details that are pertinent to understanding the embodiments of the present disclosure so as not to obscure the figures with details that will be readily apparent to those skilled in the art having the benefit of the description herein.
DETAILED DESCRIPTION
[0020] For the purpose of promoting an understanding of the principles of the disclosure, reference will now be made to the embodiment illustrated in the figures and specific language will be used to describe them. It will nevertheless be understood that no limitation of the scope of the disclosure is thereby intended. Such alterations and further modifications in the illustrated system, and such further applications of the principles of the disclosure as would normally occur to those skilled in the art are to be construed as being within the scope of the present disclosure.
[0021] The terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process or method that comprises a list of steps does not include only those steps but may include other steps not expressly listed or inherent to such a process or method. Similarly, one or more devices or sub-systems or elements or structures or components preceded by "comprises... a" does not, without more constraints, preclude the existence of other devices, sub-systems, elements, structures, components, additional devices, additional sub-systems, additional elements, additional structures, or additional components. Appearances of the phrase "in an embodiment", "in another embodiment" and similar language throughout this specification may, but not necessarily do, all refer to the same embodiment.
[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this disclosure belongs. The system, methods, and examples provided herein are only illustrative and not intended to be limiting.
[0023] In the following specification and the claims, reference will be made to a number of terms, which shall be defined to have the following meanings. The singular forms “a”, “an”, and “the” include plural references unless the context clearly dictates otherwise.
[0024] Embodiments of the present disclosure relate to a system for counting articles and detecting defects in the articles based on machine learning. The system includes a processing subsystem, hosted on a server, and configured to execute on a network to control bidirectional communications among a plurality of modules. The plurality of modules includes an article detection module, a counting module, a defect detection module, a prediction module, an identification module, and a recommendation module. The article detection module is configured to capture at least one of a video and an image of the plurality of articles moving on a conveyor belt in real-time. The article detection module is also configured to process the at least one of video and image for noise removal, background subtraction and resize of the image. Further, the article detection module is configured to provide a bounding box across a region of interest of the article to detect the article. The counting module is operatively coupled with the article detection module. The counting module is configured to count one or more regions of interest to count the number of articles moving on the conveyor belt. The defect detection module is operatively coupled with the article detection module. The defect detection module includes a visual unit for generating three-dimensional images of the article. The generated three-dimensional images are used to detect one or more defects in the article. The detected defects are stored in a repository. The prediction module is operatively coupled with the defect detection module. The prediction module is configured to predict one or more corrective measures for the defective article by comparing the detected defects with the stored defects in the repository. The prediction is based on a machine learning model collect the real-time data of the detected defects with respect to the predicted corrective measures. The prediction module is configured to send the collected data to an article making machine for enabling active learning. The article making machine is operatively coupled with the processing subsystem. The predicted corrective measures are stored in a database. The identification is module operatively coupled with the prediction module, wherein the identification module is configured to identify an anomaly of the article making machine. The identification of the anomaly is based on internet of things sensors connected to the article making machine. The recommendation module operatively coupled with the prediction module and the identification module. The recommendation module is configured to recommend a plurality of preventive measures by comparing with the predicted corrective measures stored in the database. The plurality of preventive measures is recommended in a plurality of languages. The recommendation module is also configured to provide feedback on the recommended plurality of corrective measures to the article making machine.
[0025] FIG. 1 is a block diagram representing a system (100) for counting articles and detecting defects in articles in accordance with an embodiment of the present disclosure. The system (100) includes a processing subsystem (104).The processing subsystem (104) is hosted on a server (106). The processing subsystem (104) is configured to execute on a network (108) to enable communications among a plurality of modules. In one embodiment, the server (106) may include a cloud server. In another embodiment, the server (106) may include a local server. The processing subsystem (104) is configured to execute on a network (108) to control bidirectional communications among a plurality of modules. In one embodiment, the network (108) may include a wired network such as a local area network (LAN). In another embodiment, the network may include a wireless network such as Wi-Fi, Bluetooth, Zigbee, near-field communication (NFC), infrared communication (RFID), or the like.
[0026] Further, the plurality of modules includes an article detection module (110), a counting module (112), a defect detection module (114), a prediction module (120), an identification module (126), and a recommendation module (128).
[0027] The article detection module (110) is configured to capture at least one of a video and an image of the plurality of articles (130) moving on a conveyor belt (132) in real-time. In a preferred embodiment, the plurality of articles (130) are samples of bottles (glass, aluminium, plastic, stainless steel, black, coloured and frosted). The article detection module (110) is also configured to process the at least one of video and image for noise removal, background subtraction and resize of the image. Further, the article detection module (110) is configured to provide a bounding box across a region of interest of the article to detect the article. In one embodiment, the article detection module (110) and the counting module (112) are based on a convolution neural network model. In one embodiment, the real-time data (video and images) includes data collected from the articles making machine, data related to articles, and data related to the operator of the articles making machine (122).
[0001] It must be noted that the article detection module (110) is one or more cameras to capture the at least one of a video and an image of the plurality of articles (130). Specifically, the one or more cameras must be capable of capturing high resolution images of the articles. In one embodiment, the one or more cameras may be an analog or digital still image camera, a video camera, an optical camera, a laser camera, a 3D image scanner, full spectrum camera, an infrared illuminated camera, an infrared camera, a thermal imaging camera, a stereoscopic camera, and/or any other combination of cameras or different types of cameras.
[0028] The counting module (112) is operatively coupled with the article detection module (110). The counting module (112) is configured to count one or more regions of interest to count the number of articles moving on the conveyor belt (132). In one embodiment, the counting module (112) counts the articles based on a custom counting model, wherein the custom count model counts the region of interest to count articles moving on the conveyer.
[0029] The defect detection module (114) is operatively coupled with the article detection module (110). The defect detection module (114) includes a visual unit (116) for generating three-dimensional images of the article. The generated three-dimensional images are used to detect one or more defects in the article. The detected defects are stored in a repository (118). Examples of the one or more defects includes, but is not limited to, cracks, holes and damage. Typically, the said one or more defects are not visible to a naked eye. Further, the articles may have a variety of shapes, such as cylindrical, elliptical cylindrical, other irregular shapes and the like.
[0030] The prediction module (120) is operatively coupled with the defect detection module (114). The prediction module (120) is configured to predict one or more corrective measures for the article by comparing the detected defects with the stored defects in the repository (118). The prediction is based on the machine learning model. The prediction module (120) is also configured to collect the real-time data of the detected defects with respect to the predicted corrective measures. Further, the prediction module (120) is configured to send the collected data to an article making machine (122) for enabling active learning. The article making machine (122) is operatively coupled with the processing subsystem (104). The predicted corrective measures are stored in a database (124).
[0031] The identification module (126) is operatively coupled with the prediction module (120). The identification module (126) is configured to identify an anomaly of the article making machine (122). The identification of the anomaly is based on Internet of Things (IoT) sensors (134) connected to the article making machine (122).
[0032] The recommendation module (128) is operatively coupled with the prediction module (120) and the identification module (126). The recommendation module (128) is configured to recommend a plurality of preventive measures by comparing with the predicted corrective measures stored in the database (124). The plurality of preventive measures is recommended in a plurality of languages. The recommendation module (128) is configured to provide feedback on the recommended plurality of corrective measures to the article making machine (122). In one embodiment, the recommendation module (128) is configured to recommend a plurality of preventive measures to at least one of the articles making machine (122) and an operator of the article making machine (122).
[0033] FIG. 2 is a block diagram of an exemplary embodiment of the system for counting articles and detecting defects in the articles of FIG. 1 in accordance with an embodiment of the present disclosure. Considering a non-limiting example of a bottle as the article. The manufactured bottles are placed on the conveyer belt (132) and passes in sequence under the transmission of the conveyer belt. The video or the image of the bottle moving on the conveyor is captured in real time. The image is then divided into frames and pre-processed before giving it as an input to the article detection module (110) configured with a custom model trained to track the bottles. The pre-processing stage involves noise removal, background subtraction and resizing of image to match the trained model's input. The output of the article detection module (110) involves a bounding box drawn across the region of interest (ROI) of bottles. In one embodiment, the article detection module (110) is configured to generate a three-dimensional model by using at least one of image sensors (216), X-rays (218), and HoloLens (220).
[0034] In one embodiment, the output from the article detection module (110) is given as the input to the counting module (112) which includes of a custom counting algorithm. The custom counting algorithm tracks the ROI, then takes a count of ROI and gives a count of the number of bottles moving on the conveyor belt (132). In one embodiment, the defects in the bottles may occur due to faulty article making machine (122) or wrong article making machine (122) settings. The internet of things (IoT) (134) sensors is used to identify the status of the article making machine. The status of the article making machine (122) includes the working condition of the article making machine (122). The anomaly in the article making machine (122) is identified by the IoT sensors (134). The recommendation module (128) provides recommendation to prevent defects in the bottles in future. There are many different types of defects that can occur in bottles and may not be visible to bare eyes hence the defect detection module (114) is needed. In one embodiment, the defect detection module (114) detects defects by using HoloLens. In one embodiment, a classifier (212) includes a plurality of lens which may be used for classifying the defects present in the bottle at the time of manufacturing. Upon defect detection, the prediction module (120) recommends corrective measure to the machine. Also, preventive measures are recommended for the predicted defect which may occur in the near future. At the same time, the feedback on the prediction module (120) collects real time data (202) and fed it back to the article making machine (122) to make it more robust. The prediction module (120) actively learns at every instance. In one order to make the system robust, an active learning mode is provided by adding feedback loop (214) which facilitates the system (100) to continuously learn. The active learning is conducted based on the inputted program details (204) which includes type of job (bottle), speed of the conveyer belt, size and shape of the bottle. Also, learning weights are taking from the machine learning model for active learning (208). A multilinguistic feature is provided by using multilinguistic recommender engine (210) for making this system usable by many users.
[0035] FIG. 3 is a block diagram of a computer or a server (300) for the system (100) for counting articles and detecting defects in articles in accordance with an embodiment of the present disclosure. The server (300) includes a processor(s) (302), and memory (306) operatively coupled to the bus (304).
[0036] The processor(s) (302), as used herein, means any type of computational circuit, such as, but not limited to, a microprocessor, a microcontroller, a complex instruction set computing microprocessor, a reduced instruction set computing microprocessor, a very long instruction word microprocessor, an explicitly parallel instruction computing microprocessor, a digital signal processor, or any other type of processing circuit, or a combination thereof.
[0037] The bus (304) as used herein refers to be internal memory channels or computer network that is used to connect computer components and transfer data between them. The bus (304) includes a serial bus or a parallel bus, wherein the serial bus transmits data in a bit-serial format and the parallel bus transmits data across multiple wires. The bus (304) as used herein, may include but not limited to, a system bus, an internal bus, an external bus, an expansion bus, a frontside bus, a backside bus, and the like.
[0038] The memory (306) includes a plurality of subsystems and a plurality of modules stored in the form of an executable program which instructs the processor (302) to perform the method steps illustrated in FIG. 1. The memory (306) is substantially similar to the system (100) of FIG.1. The memory (306) has submodules: an article detection module (110), a counting module (112), a defect detection module (114), a prediction module (120), an identification module (126), and a recommendation module (128).
[0039] The article detection module (110) is configured to capture at least one of a video and an image of the plurality of articles (130) moving on a conveyor belt (132) in real-time. The article detection module (110) is also configured to process the at least one of video and image for noise removal, background subtraction and resize of the image. Further, article detection module (110) is configured to provide a bounding box across a region of interest of the article to detect the article.
[0040] The counting module (112) is operatively coupled with the article detection module (110). The counting module (112) is configured to count one or more regions of interest to count the number of articles moving on the conveyor belt (132).
[0041] The defect detection module (114) is operatively coupled with the article detection module (110). The defect detection module (114) includes a visual unit (116) for generating three-dimensional images of the article. the generated three-dimensional images are used to detect one or more defects in the article. The detected defects are stored in a repository (118).
[0042] The prediction module (120) is operatively coupled with the defect detection module (114). The prediction module (120) is configured to predict one or more corrective measures for the article by comparing the detected defects with the stored defects in the repository (118). The prediction is based on the machine learning model. The prediction module (120) is also configured to collect the real-time data of the detected defects with respect to the predicted corrective measures. Further, the prediction module (120) is configured to send the collected data to an article making machine (122) for enabling active learning. The article making machine (122) is operatively coupled with the processing subsystem (104). The predicted corrective measures are stored in a database (124).
[0043] The identification module (126) is operatively coupled with the prediction module (120). The identification module (126) is configured to identify an anomaly of the article making machine. The identification of the anomaly is based on internet of things sensors (134) connected to the article making machine (122).
[0044] The recommendation module (128) is operatively coupled with the prediction module (120) and the identification module (126). The recommendation module (128) is configured to recommend a plurality of preventive measures by comparing with the predicted corrective measures stored in the database (124). The plurality of preventive measures is recommended in a plurality of languages. The recommendation module (128) is configured to provide feedback on the recommended plurality of corrective measures to the article making machine (122).
[0045] FIG. 4 is a flow chart representing steps involved in a method (400) for counting articles and detecting defects in articles in accordance with an embodiment of the present disclosure. The method (400) includes capturing, by an article detection module of a processing subsystem, capture at least one of a video and an image of the plurality of articles moving on a conveyor belt in real-time in step (402).
[0046] The method (400) also includes processing, by the article detection module of the processing subsystem, the at least one of video and image for noise removal, background subtraction and resize of the image in step (404).
[0047] Further, the method (400) includes providing, by the article detection module of the processing subsystem, a bounding box across a region of interest of the article to detect the article in step (406). In one embodiment, the article detection module is based on a convolution neural network model. The method also include generating, a three-dimensional model by using at least one of machine vision cameras, X-rays, and HoloLens.
[0048] Furthermore, the method (400) includes counting, by a counting module of the processing subsystem, one or more regions of interest to count the number of articles moving on the conveyor belt instep (408). The method also includes the counting, the articles based on a custom counting model, wherein the custom count model counts the region of interest to count articles moving on the conveyer. In one embodiment, the counting module is based on a convolution neural network model. The method also includes counting, one or more defective articles and one or more non-defective articles.
[0049] Furthermore, the method (400) includes generating, by a visual unit of a defect detection module of the processing subsystem, three-dimensional images of the article in step (410).
[0050] Furthermore, the method (400) includes detecting, by the defect detection module of the processing subsystem, one or more defects in the article in step (412).
[0051] Furthermore, the method (400) includes storing by a repository of the defect detection module of the processing subsystem, the detected defects in step (414).
[0052] Furthermore, the method (400) includes predicting, by a prediction module of the processing subsystem, one or more corrective measures for the article by comparing the detected defects with the stored defects in the repository, wherein the predicting is based on a machine learning model in step (416).
[0053] Furthermore, the method (400) includes collecting, by the prediction module of the processing subsystem, the real-time data of the detected defects with respect to the predicted corrective measures in step (418). The method also includes collecting, the real-time data includes data collected from the articles making machine, the real-time data related to articles, and data related to the operator of the articles making machine.
[0054] Furthermore, the method (400) includes sending, by the prediction module of the processing subsystem, the collected data to an article making machine for enabling active learning, wherein the article making machine is operatively coupled with the processing subsystem in step (420) .
[0055] Furthermore, the method (400) includes storing, by the prediction module of the processing subsystem, the predicted corrective measures in a database in step (422).
[0056] Furthermore, the method (400) includes identifying, by an identification module of the processing subsystem, an anomaly of the article making machine, and wherein the identification of the anomaly is based on internet of things sensors connected to the article making machine in step (424).
[0057] Furthermore, the method (400) includes recommending, by a recommendation module of the processing subsystem, a plurality of preventive measures by comparing with the predicted corrective measures stored in the database, wherein the plurality of preventive measures is recommended in a plurality of languages in step (426). The method also includes recommending a plurality of preventive measures to at least one of the articles making machine and an operator of the article making machine.
[0058] Furthermore, the method (400) includes providing, by the recommendation module of the processing subsystem, feedback on the recommended plurality of corrective measures to the article making machine in step (428).
[0059] Various embodiments of the present disclosure enable an automatic system for counting articles and detecting defects in articles. The system disclosed in the present disclosure provides a real time insight about the cause of defects. to the manufacturers. The system disclosed in the present disclosure suggests a corrective measure for the defect and that may be immediately taken to minimize rejections of the articles. Further, the system in the present disclosure avoids human error. The system facilitates counting of articles and defect detection in real-time by using machine learning. The system facilitate recommendation in various languages so as the multiple users are able to use.
[0060] While specific language has been used to describe the disclosure, any limitations arising on account of the same are not intended. As would be apparent to a person skilled in the art, various working modifications may be made to the method in order to implement the inventive concept as taught herein.
[0061] The figures and the foregoing description give examples of embodiments. Those skilled in the art will appreciate that one or more of the described elements may well be combined into a single functional element. Alternatively, certain elements may be split into multiple functional elements. Elements from one embodiment may be added to another embodiment. For example, order of processes described herein may be changed and are not limited to the manner described herein. Moreover, the actions of any flow diagram need not be implemented in the order shown; nor do all of the acts need to be necessarily performed. Also, those acts that are not dependent on other acts may be performed in parallel with the other acts. The scope of embodiments is by no means limited by these specific examples.
, C , Claims:1. A system (100) for counting articles and detecting defects in the articles based on a machine learning model comprising:
a processing subsystem (104), hosted on a server (106), and configured to execute on a network (108) to control bidirectional communications among a plurality of modules comprising:
an article detection module (110) configured to:
capture at least one of a video and an image of the plurality of articles (130) moving on a conveyor belt (132) in real-time;
process the at least one of video and image for noise removal, background subtraction and resize of the image; and
provide a bounding box across a region of interest of the article to detect the article;
a counting module (112) operatively coupled with the article detection module (110), wherein the counting module (112) is configured to count one or more regions of interest to count the number of articles moving on the conveyor belt (132);
a defect detection module (114) operatively coupled with the article detection module (110), wherein the defect detection module (114) comprises a visual unit (116) for generating three-dimensional images of the article,
wherein the generated three-dimensional images are used to detect one or more defects in the article, and
wherein the detected defects are stored in a repository (118);
a prediction module (120) operatively coupled with the defect detection module (114), wherein the prediction module (120) is configured to:
predict one or more corrective measures for the article by comparing the detected defects with the stored defects in the repository (118), wherein the prediction is based on the machine learning model;
collect the real-time data of the detected defects with respect to the predicted corrective measures; and
send the collected data to an article making machine (122) for enabling active learning, wherein the article making machine (122) is operatively coupled with the processing subsystem (104), and
wherein the predicted corrective measures are stored in a database (124);
an identification module (126) operatively coupled with the prediction module (120), wherein the identification module (126) is configured to identify an anomaly of the article making machine, and wherein the identification of the anomaly is based on internet of things sensors (134) connected to the article making machine (122); and
a recommendation module (128) operatively coupled with the prediction module (120) and the identification module (126), wherein the recommendation module (128) is configured to:
recommend a plurality of preventive measures by comparing with the predicted corrective measures stored in the database, wherein the plurality of preventive measures is recommended in a plurality of languages; and
provide feedback on the recommended plurality of corrective measures to the article making machine (122).
2. The system (100) as claimed in claim 1, wherein the article detection module (110) and the counting module (112) are based on a convolution neural network model.
3. The system (100) as claimed in claim 1, wherein the article detection module (110) is configured to generate a three-dimensional model by using at least one of machine vision cameras, X-rays, and HoloLens.
4. The system (100) as claimed in claim 1, wherein the counting module (112) counts the articles based on a custom counting model, wherein the custom count model counts the region of interest to count articles moving on the conveyor belt (132).
5. The system (100) as claimed in claim 1, wherein the counting module (112) is configured to count one or more defective articles and one or more non-defective articles.
6. The system (100) as claimed in claim 1, wherein the recommendation module (128) is configured to recommend a plurality of preventive measures to at least one of the articles making machine (122) and an operator of the article making machine (122).
7. The system (100) as claimed in claim 1, wherein the real-time data comprises data collected from the articles making machine, data related to articles, and data related to the operator of the articles making machine (122).
8. A method (400) for counting articles and detecting defects in an article comprises:
capturing, by an article detection module of a processing subsystem, capture at least one of a video and an image of the plurality of articles moving on a conveyor belt in real-time; (402)
processing, by the article detection module of the processing subsystem, the at least one of video and image for noise removal, background subtraction and resize of the image; (404)
providing, by the article detection module of the processing subsystem, a bounding box across a region of interest of the article to detect the article; (406)
counting, by a counting module of the processing subsystem, one or more regions of interest to count the number of articles moving on the conveyor belt; (408)
generating, by a visual unit of a defect detection module of the processing subsystem, three-dimensional images of the article; (410)
detecting, by the defect detection module of the processing subsystem, one or more defects in the article; (412)
storing by a repository of the defect detection module of the processing subsystem, the detected defects; (414)
predicting, by a prediction module of the processing subsystem, one or more corrective measures for the article by comparing the detected defects with the stored defects in the repository, wherein the predicting is based on a machine learning model; (416)
collecting, by the prediction module of the processing subsystem, the real-time data of the detected defects with respect to the predicted corrective measures; (418)
sending, by the prediction module of the processing subsystem, the collected data to an article making machine for enabling active learning, wherein the article making machine is operatively coupled with the processing subsystem; (420)
storing, by the prediction module of the processing subsystem, the predicted corrective measures in a database; (422)
identifying, by an identification module of the processing subsystem, an anomaly of the article making machine, and wherein the identification of the anomaly is based on internet of things sensors connected to the article making machine; (424)
recommending, by a recommendation module of the processing subsystem, a plurality of preventive measures by comparing with the predicted corrective measures stored in the database, wherein the plurality of preventive measures is recommended in a plurality of languages; (426) and
providing, by the recommendation module of the processing subsystem, feedback on the recommended plurality of corrective measures to the article making machine. (428)

Dated this 07th day of August 2023

Signature

Jinsu Abraham
Patent Agent (IN/PA-3267)
Agent for the Applicant