Description:TECHNICAL FIELD
[001] The present disclosure relates generally to the field of inventory management systems for industrial applications. More particularly, the present disclosure relates to a system and method utilising computer vision and machine learning techniques for automated detection, counting, measurement, and inventory verification of pipes in storage environments.

BACKGROUND
[002] Background description includes information that may be useful in understanding the present invention. It is not an admission that any of the information provided herein is prior art or relevant to the presently claimed disclosure, or that any publication specifically or implicitly referenced is prior art.
[003] Pipe inventory management in manufacturing and distribution industries has traditionally relied on manual counting and measurement methods. These manual processes are time-consuming, susceptible to human error, and often lead to inaccurate inventory records. Such inaccuracies can cause significant operational and financial issues, including incorrect shipments, production delays, and inventory shrinkage.
[004] In early practices, inventory management includes physical counting of pipes by workers. However, manual counting is inherently error-prone and often results in miscounts that propagate through inventory records, disrupting production planning and order fulfillment. Additionally, this approach includes high labour costs, delivers inconsistent results, and lacks real-time visibility of inventory status.
[005] To improve traceability, some industries adopted barcode or RFID tagging of pipe bundles. While this provided better tracking, it still required manual tagging and scanning operations, creating inefficiencies and bottlenecks. Moreover, such tags are prone to damage or loss in harsh industrial environments, leading to further inventory inaccuracies.
[006] Semi-automated counting solutions have been developed using basic computer vision techniques to identify and count pipes. However, these methods have struggled with practical challenges such as variable lighting conditions, reflective surfaces, overlapping pipes, and, particularly the accurate identification of nested pipe arrangements. These limitations often resulted in insufficient accuracy for commercial use, especially when handling pipes of varying diameters.
[007] Conventional image processing solutions were also introduced, but proved unreliable in complex environments. Overlapping objects, shadows, industrial debris, and inconsistent pipe orientations frequently caused false detections or counting errors. These systems often required tightly controlled environments to function effectively, reducing their practical value.
[008] Some modern inventory management approaches integrated machine learning-based object recognition to improve detection accuracy. However, these solutions were generally not optimised for the specific characteristics of pipes and often depended on cloud-based computing resources. Such reliance introduced issues related to latency, network availability, and increased operational costs, making them less suitable for industrial environments where edge computing is preferred.
[009] Existing solutions suffer from critical drawbacks, including an inability to accurately detect and count nested pipe arrangements, lack of real-time verification between existing inventory and outgoing shipments, poor adaptability to challenging industrial environments, and dependence on high computing resources that increase implementation complexity and costs.
[010] Therefore, there is a need for a technology that provides a reliable and automated solution for pipe inventory management, capable of accurately counting pipes, including nested arrangements, verifying inventory records against shipments, and operating efficiently in industrial environments.

OBJECTS OF THE PRESENT DISCLOSURE
[011] A general object of the present disclosure is to provide a system and method that reduces labour requirements for pipe inventory management.
[012] An object of the present disclosure is to provide a system and method that improves inventory accuracy in industrial environments.
[013] Another object of the present disclosure is to provide a system and method capable of accurately counting pipes arranged in nested configurations.
[014] Another object of the present disclosure is to provide a system and method that prevents unauthorised removal of pipes through inventory cross-verification.
[015] Another object of the present disclosure is to provide a system and method that functions reliably in areas with limited or no network connectivity.
[016] Another object of the present disclosure is to provide a system and method that offers a detailed characterisation of pipes based on diameter and wall thickness.

SUMMARY
[017] Aspects of the present disclosure relate to the field of inventory management systems for industrial applications. More particularly, the present disclosure relates to a system and method utilising computer vision and machine learning techniques for automated detection, counting, measurement, and inventory verification of pipes in storage environments.
[018] The present disclosure provides a pipe counting system designed for automated inventory management within a pre-defined pipe storage area. The system includes at least one image acquisition unit configured to capture images of the storage area containing multiple pipes. These captured images are processed by a processing unit that operates in conjunction with a memory storing executable instructions to carry out the functions described herein. Upon receiving the acquired images from the image acquisition unit, the processing unit applies a series of computer vision techniques to extract meaningful visual data from the images. These techniques include specific preprocessing operations such as contrast adjustment, histogram equalisation, and convolution-based edge operations to ensure accurate detection of the pipes regardless of variable lighting conditions or environmental factors within industrial settings.
[019] Following pre-processing, the system detects and identifies individual pipes present in the images. The system is capable of recognising not only individual pipes but also nested pipes arranged in complex configurations. For this purpose, the processing unit applies a deep learning model based on YOLOv11, which enables the identification of pipe structures within the images with high accuracy. Subsequent to detection, the system evaluates geometric properties of each identified pipe. These properties include diameter and wall thickness, which are determined through a combination of contour analysis techniques and pixel-to-millimeter calibration. The calibration is performed using reference objects that are present within the captured images, ensuring precise measurements for accurate inventory records.
[020] The system further distinguishes between inner contours and outer contours of nested pipes through contour analysis. This differentiation is essential for accurately measuring geometric properties and preventing miscounts where multiple pipes overlap or are arranged concentrically. The camera module integrated with a Raspberry Pi continuously monitors the storage rack and counts the pipes present. Upon completion of detection and evaluation, the system automatically updates the rack inventory database with the current pipe count. The database maintains real-time records of pipe quantities associated with their respective storage rack locations..
[021] The processing unit analyzes the captured images and generates output data containing the accurate count of pipes present in the monitored storage rack. This count data is automatically stored in the database and provided to users through a computing device, offering real-time visibility into current rack inventory levels.
[022] Another aspect of the present disclosure pertains to a method for counting pipes in storage racks and during shipment operations. The method includes acquiring images of storage areas using an integrated camera module, processing these images using computer vision techniques including contrast adjustment, histogram equalisation, and convolution-based edge operations. The method continues with detecting and identifying individual and nested pipes using a YOLOv11-based deep learning model. During normal operations, the system continuously monitors storage racks and updates the rack inventory database with current pipe counts. During shipment operations, the system captures images of pipes being shipped, performs the same analysis to count the pipes, and records this data in a separate shipment database. The method enables discrepancy detection by comparing total production quantities against the sum of rack inventory and shipment records, thereby identifying any unauthorized removal or counting errors.
[023] In an aspect, through this combination of computer vision, deep learning, and geometric analysis techniques, the present disclosure offers an accurate, automated, and reliable solution for pipe inventory management in industrial environments.
[024] Various objects, features, aspects, and advantages of the inventive subject matter will become more apparent from the following detailed description of preferred embodiments, along with the accompanying drawing figures in which like numerals represent components.

BRIEF DESCRIPTION OF THE DRAWINGS
[025] The accompanying drawings are included to provide a further understanding of the present disclosure and are incorporated in and constitute a part of this specification. The drawings illustrate exemplary embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure.
[026] FIG. 1 illustrates an exemplary architecture of proposed automated pipe counting system for inventory management, in accordance with an embodiment of the present disclosure.
[027] FIG. 2 illustrates an exemplary flow chart to depict working of proposed automated pipe counting system for inventory management, in accordance with an embodiment of the present disclosure.
[028] FIGs. 3A and 3B illustrate exemplary views of an original image and a corresponding contour image of pipes, in accordance with an embodiment of the present disclosure.
[029] FIG. 4 illustrates bounding box identification of pipes within a predefined area using a YOLOv11 model, in accordance with an embodiment of the present disclosure.
[030] FIGs. 5A-5H illustrate exemplary views of circle extraction and contour detection process, showing identification of circular patterns in nested pipe arrangements, in accordance with an embodiment of the present disclosure.
[031] FIGs. 6A-6E illustrate exemplary views of the diameter and wall thickness measurement process, showing detection of pipe circles, in accordance with an embodiment of the present disclosure.
[032] FIG. 7 illustrates an exemplary flow diagram of the real-time detection pipeline, from image acquisition to result visualisation, in accordance with an embodiment of the present disclosure.
[033] FIG. 8 illustrates an exemplary flow chart of a method for counting pipes for inventory management, in accordance with an embodiment of the present disclosure.
[034] FIG. 9 illustrates a block diagram of an example computer system in which or with which embodiments of the present disclosure may be implemented.

DETAILED DESCRIPTION
[035] The following is a detailed description of embodiments of the disclosure depicted in the accompanying drawings. The embodiments are in such detail as to clearly communicate the disclosure. However, the amount of detail offered is not intended to limit the anticipated variations of embodiments; on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the present disclosures as defined by the appended claims.
[036] Embodiments explained herein relate to the T the field of inventory management systems for industrial applications. More particularly, the present disclosure relates to a system and method utilising computer vision and machine learning techniques for automated detection, counting, measurement, and inventory verification of pipes in storage environments.
[037] Referring to FIG. 1, an exemplary block diagram (100) of proposed automated pipe counting system (102) for inventory management is disclosed. The system (102) is used to automatically detect, count, and measure pipes in real-time, enabling accurate and efficient inventory tracking in industrial settings. The system (102) includes at least one image acquisition unit (104) configured to acquire images of a pre-defined pipe storage area. The pre-defined pipe storage area can be a section within a warehouse, storage yard, manufacturing facility, or a shipment container where pipes are systematically arranged or stacked for storage or transportation. This allows the system to function not only in static storage environments but also during shipment operations to ensure accurate tracking and verification of inventory. The image acquisition unit (104) may include one or more cameras or imaging devices installed at fixed positions or mounted on movable platforms, capable of capturing high-resolution images that cover the entire pipe area.
[038] In addition, the system (102) includes a processing unit (106) operatively coupled to the image acquisition unit (102) and a memory (108). The processing unit (106) may be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, logic circuitries, and/or any devices that manipulate data based on operational instructions. Among other capabilities, the processing unit (106) may be configured to fetch and execute computer-readable instructions stored in the memory (108) of the system (102). The memory (108) may store one or more computer-readable instructions or routines, which may be fetched and executed for counting pipes automatically. The memory (108) may include any non-transitory storage device, including, for example, volatile memory such as Random-Access Memory (RAM), or non-volatile memory such as Erasable Programmable Read-Only Memory (EPROM), flash memory, and the like.
[039] In an embodiment, the processing unit (106) may also include an interface(s) (114). The interface(s) (114) may include a variety of interfaces, for example, a variety of interfaces, interfaces for data input and output devices, referred to as I/O devices, storage devices, and the like. The interface(s) (114) may facilitate communication of the system (102) with various devices coupled to it. The interface(s) (114) may also provide a communication pathway for one or more components of the system (102). Examples of such components include, but are not limited to, processing engine(s) (116) and a database (118). The database (118) may include data that is either stored or generated as a result of functionalities implemented by any of the components of the processing engine(s) (116).
[040] In an embodiment, the processing engine(s) (116) may be implemented as a combination of hardware and programming (for example, programmable instructions) to implement one or more functionalities of the processing engine(s) (116). In the examples described herein, such combinations of hardware and programming may be implemented in several different ways. For example, the programming for the processing engine(s) (116) may be processor executable instructions stored on a non-transitory machine-readable storage medium, and the hardware for the processing unit (106) may comprise a processing resource (for example, one or more processors), to execute such instructions. In the present examples, the machine-readable storage medium may store instructions that, when executed by the processing resource, implement the processing engine(s) (208).
[041] In such examples, the system (102) may include the machine-readable storage medium storing the instructions and the processing resource to execute the instructions, or the machine-readable storage medium may be separate but accessible to the system (102) and the processing resource. In other examples, the processing engine(s) (116) may be implemented by an electronic circuitry. The processing engine(s) (116) may include a pre-processing module (122), a pipe detection and identification module (124), a geometric analysis and calibration module (126), an inventory management and discrepancy detection module (128) and other module(s) (130). The other module(s) (130) may implement functionalities that supplement applications/functions performed by the processing engine(s) (116).
[042] In an embodiment, the pre-processing module (122) is configured for handling and analysing the images captured by the image acquisition unit (104). Upon receiving these images, the processing unit (106) applies computer vision techniques to identify and distinguish individual pipes present within the captured scene. This includes the detection of nested pipes, which are pipes placed inside one another, often making the counting and measurement task more complex. To enhance the quality of the images for accurate detection, the processing unit (106) first performs preprocessing operations such as contrast adjustment to enhance visual clarity, histogram equalisation to normalise brightness and contrast across the image, and convolution-based edge operations to highlight the outlines and boundaries of the pipes. These pre-processing steps assist in bringing out the structural features of the pipes, especially when they are stacked or partially obscured.
[043] Following pre-processing, the pipe detection and identification module (124) utilises a You Only Look Once, version 11 (YOLOv11) based deep learning model to accurately detect and identify the individual pipes. YOLOv11 is a fast and efficient object detection algorithm that can identify multiple objects in a single frame with high precision. By applying this model to the enhanced images, the system (102) is able to detect not only clearly visible pipes but also those nested within other pipes, ensuring robust and automated pipe counting for inventory management.
[044] In an embodiment, the geometric analysis and calibration module (126) is configured to evaluate one or more geometric properties of the identified pipes, including the diameter and wall thickness. This evaluation is performed by applying contour analysis techniques to the segmented images of the pipes. To accurately convert image-based measurements into real-world dimensions, the processing unit uses pixel-to-millimeter calibration. This calibration process relies on one or more reference objects of known dimensions that are captured within the same images. By analysing the size of these reference objects in pixels, the system (102) establishes a scale for translating pixel measurements into millimeters. Using this calibrated scale, the system then computes the actual diameter and wall thickness of each identified pipe, enabling accurate categorisation and quantification for inventory and quality control purposes.
[045] Further, the geometric analysis and calibration module (126) is configured to identify inner contours and outer contours of the nested pipes by performing contour analysis on the detected pipes. The contour analysis enables the system (102) to distinguish between the internal and external boundaries of each pipe, which is essential for accurately determining geometric properties such as diameter and wall thickness. By analysing these contours, the system (102) ensures precise measurements even in cases where pipes are stored within one another.
[046] In an embodiment, the inventory management and discrepancy detection module (128) is configured to generate an inventory dataset based on the count of the identified pipes within the pre-defined pipe storage area. This inventory dataset includes an association between the number of pipes detected and their corresponding location in the storage area. The processing unit (106) is also configured to compare the generated inventory dataset with existing inventory records stored in the memory (108). This comparison enables the system (102) to detect discrepancies, such as unauthorised removal of pipes, by identifying any differences between the current and previously stored inventory data.
[047] Further, the inventory management and discrepancy detection module (128) generates an output that includes current inventory status, shipment verification results, and any detected discrepancies based on the comparison between the generated inventory dataset and stored inventory records. This output is displayed on a computing device (110), which allows authorised personnel to view and assess the real-time inventory status. The computing device (110) is communicatively coupled to the processing unit (106) through a communication unit (112), enabling seamless data exchange and remote monitoring capabilities. The computing device (110) can be a desktop computer, laptop, tablet, or mobile device used by inventory managers, warehouse staff, or supervisors to view the generated inventory data, shipment verification results, and discrepancies. This serves as a user interface for interacting with the system (102).
[048] The communication unit (112) can be a wireless module such as Wi-Fi, Bluetooth, or a mobile communication module, or a wired interface such as Ethernet or USB. This facilitates data transmission between the processing unit (106) and the computing device (110), ensuring reliable communication and synchronisation of information.
[049] Referring to FIG. 2, an exemplary flow chart (200) depicts the working of the proposed automated pipe counting system (102) for inventory management. At step (202), images of storage racks or shipments are captured by the image acquisition unit (104). These captured images are pre-processed at step (204) to enhance the features relevant for pipe detection. Following this, at step (206), a hybrid approach is used to detect pipes, which combines YOLOv11-based deep learning with traditional computer vision techniques. Once the pipes are detected, their dimensions, such as diameter and wall thickness, are calculated at step (208) using contour analysis. The system (102) also enables continuous monitoring. At step (210), the extracted data is stored in a dual-database (118). This includes a rack inventory database, as shown in step (218), and a shipment verification database, as shown in step (220). The data from both these databases is transmitted to step (212), where cross-verification is carried out by comparing the expected inventory with the actual current inventory. Based on this comparison, results are displayed at step (214), showing inventory status and generating alerts for any discrepancies. Further, at step (216), local processing is performed using edge computing capabilities, reducing dependency on network connectivity.
[050] Referring to FIGs. 3A and 3B, exemplary views of an original image and a corresponding contour image of pipes are depicted. FIG. 3A shows raw captured image of the pipes as seen by the image acquisition unit (104), while FIG. 3B illustrates the result after applying contour analysis, highlighting the pipe boundaries for further processing and measurement.
[051] Referring to FIG. 4, bounding box identification of pipes within the predefined area using a YOLOv11 model is depicted. The YOLOv11 model detects and outlines individual pipes by drawing bounding boxes around them. Each bounding box is associated with a confidence score, such as 0.3, 0.9, 0.95, 0.96, 0.91, or 0.973, which represents the likelihood that the identified object is indeed a pipe. Higher scores indicate more accurate identification, supporting reliability of the system (102) in recognising and isolating pipes within the image frame captured from the storage or shipment area.
[052] Referring to FIGs. 5A-5H, exemplary views of circle extraction and contour detection process are illustrated, demonstrating the identification of circular patterns in nested pipe arrangements. FIG. 5A shows the original captured image containing multiple pipes. FIG. 5B highlights all valid and complete pipe circles detected within the image. The subsequent figures illustrate individual pipe detections with their respective diameters: FIG. 5C indicates Pipe 1 with a diameter of 336 pixels, FIG. 5D indicates Pipe 2 with a diameter of 384 pixels, FIG. 5E shows Pipe 3 with a diameter of 494 pixels, FIG. 5F depicts Pipe 4 with a diameter of 540 pixels, FIG. 5G shows Pipe 5 with a diameter of 638 pixels, and FIG. 5H identifies Pipe 6 with a diameter of 680 pixels. These facilitate accurate contour analysis and size estimation of pipes for inventory tracking and classification.
[053] Referring to FIGs. 6A-6E, exemplary views illustrate the process of measuring the diameter and wall thickness of pipes based on the detection of pipe circles. FIG. 6A shows the original image containing multiple pipes in a nested arrangement. FIG. 6B highlights uniquely identified pipe circles used for subsequent measurement. The wall thickness of each pipe is derived by analysing the separation between the inner and outer contours of each detected circle. As shown in the result summary, the system detected six valid complete pipes, and among them, sample calculations for three pipes are displayed. FIG. 6C discloses Pipe 1, where the outer and inner diameters are 384 pixels and 336 pixels, respectively, resulting in a wall thickness of 24.0 pixels. FIG. 6D discloses Pipe 2, with an outer diameter of 540 pixels and an inner diameter of 494 pixels, giving a wall thickness of 23.0 pixels. FIG. 6E illustrates Pipe 3, where the outer and inner diameters are 680 pixels and 638 pixels, respectively, with a calculated wall thickness of 21.0 pixels. These visualisations assist in accurately estimating pipe geometry parameters, which are critical for inventory categorisation, verification, and tracking.
[054] Referring to FIG. 7, an exemplary flow diagram (700) of the real-time detection pipeline, from image acquisition to result visualisation, is disclosed. At step (702), the process begins with a user initiating image capture via a web interface through the user interface. At step (704), a signal is sent to the Raspberry Pi hardware (i.e. processing unit (106), which at step (706) activates the camera (i.e. image acquisition unit (104)) to capture the image. At step (708), the captured image undergoes an initial quality assessment to verify its usability.
[055] At step (710), the image pre-processing is initiated, where the image is converted to 8-bit grayscale. At step (712), contrast adjustment is applied to improve visual clarity, followed by histogram equalisation at step (714) to balance brightness and contrast. At step (716), normalisation is performed to standardise the image values. At step (718), Gaussian blurring with a radius of 2 is used to smooth the image. At step (720), custom convolution is applied to enhance edge features. At step (722), red channel thresholding is applied to isolate relevant regions.
[056] Continuing further, the processed image is fed into a detection engine using a YOLOv11-based model. At step (724), YOLOv11 performs object detection to identify individual pipes. At step (726), the detected regions are classified, and at step (728), bounding boxes are generated around the detected pipes, and further, at step (730), regions of interest are extracted for further analysis.
[057] Continuing further, step (732) receives data from step (730) for nested pipe analysis, performs contour extraction, and identifies sample circle points at step (734). At step (736), a Hough circle transform is applied to detect circular shapes. At step (738), filtering is performed based on radius and completeness to ensure valid circle detection. At step (740), edge type differentiation is carried out to distinguish between inner and outer contours.
[058] Continuing further, step (742) receives data from step (740) for measurement and validation, and diameter of the pipe is measured based on the detected contours. Step (744) includes converting the pixel-based measurements into millimeter-scale values using known reference objects. At step (746), confidence scores are assigned to each measurement based on detection clarity and algorithmic certainty. Further, at step (748), the measured parameters are validated by comparing them against predefined standard specifications.
[059] Continuing from step (748), step (750) receives the processed data for further database operations and cross-verification. The system stores each processed pipe record in two databases: the Rack Inventory Database (RackDB) and the Shipment Verification Database (ShipDB). In step (752), pipe metadata, such as pipe ID, timestamp, location, and detected attributes, is appended to each entry. Step (754) performs cross-verification by comparing the detected pipe parameters against stored specifications or expected values. If mismatches or anomalies are identified, they are flagged at step (756) for further action.
[060] In step (758), these results are visualized through the user interface. This includes real-time updates on the dashboard and alerts generated from flagged discrepancies in step (756), as shown in step (760). These alerts enable warehouse personnel or supervisors to immediately take corrective measures. This complete workflow facilitates an automated, accurate, and real-time solution for pipe detection, measurement, inventory logging, and verification, ensuring operational efficiency and data integrity..
[061] Referring to FIG. 8, an exemplary flow chart of a method (800) for counting pipes for inventory management is disclosed. The method (800) begins at step (802), by acquiring, by at least one image acquisition unit (104), images of a pre-defined pipe storage area containing pipes. The pre-defined pipe storage area can be a section within a warehouse, storage yard, manufacturing facility, or a shipment container where pipes are systematically arranged or stacked for storage or transportation.
[062] Continuing further at step (804), the method (800) includes receiving, by a processing unit (106), the images acquired by the image acquisition unit (104).
[063] Continuing further at step (806), the method (800) includes applying, by the processing unit (106), computer vision techniques to the received images. Applying the computer vision techniques comprises performing contrast adjustment, histogram equalisation, and convolution-based edge detection prior to detecting the pipes.
[064] Continuing further at step (808), the method (800) includes detecting and identifying, by the processing unit (106), individual pipes within the images, including nested pipes. Detecting and identifying the individual pipes, comprising nested pipes, is performed by applying the YOLOv11-based deep learning model to the pre-processed images.
[065] Continuing further at step (810), the method (800) includes evaluating, by the processing unit (106), one or more geometric properties of the identified pipes. The one or more geometric properties comprise diameter and wall thickness, and are evaluated by performing contour analysis and pixel-to-millimeter calibration based on reference objects present in the images. Performing the pixel-to-millimeter calibration comprises using one or more reference objects captured within the images to determine the diameter and the wall thickness of the identified pipes.
[066] Continuing further at step (812), the method (800) includes identifying, by the processing unit (106), inner contours and outer contours of the nested pipes based on the contour analysis to differentiate between inner and outer pipe edges. Applying the contour analysis to the detected pipes comprises differentiating between the inner contours and the outer contours of the nested pipes for determining geometric properties.
[067] Continuing further at step (814), the method (800) includes generating, by the processing unit (106), an inventory dataset based on a count of the identified pipes in the pre-defined pipe storage area. The inventory dataset includes an association between the count and the pre-defined pipe storage area.
[068] Continuing further at step (816), the method (800) includes comparing, by the processing unit (106), the generated inventory dataset with inventory records stored in a memory to detect unauthorised removal of the pipes from the pre-defined pipe storage area. For instance, the processing unit (106) enables discrepancy detection by comparing total production quantities against sum of rack inventory and shipment records, thereby identifying any unauthorized removal or counting errors
[069] Continuing further at step (818), the method (800) includes generating, by the processing unit (106), output comprising current inventory status, shipment verification results, and detected discrepancies based on the comparison. The generated output is displayed on a computing device (110), which allows authorised personnel to view and assess the real-time inventory status. The computing device (110) is communicatively coupled to the processing unit (106) through a communication unit (112), enabling seamless data exchange and remote monitoring capabilities.
[070] FIG. 9 illustrates a block diagram of an exemplary computer system (900) in which or with which embodiments of the present disclosure may be implemented.
[071] As shown in FIG. 9, the computer system (900) may include an external storage device (910), a bus (920), a main memory (930), a read-only memory (940), a mass storage device (950), communication port(s) (960), and a processor (970). A person skilled in the art will appreciate that the computer system (900) may include more than one processor and communication ports. The processor (970) may include various modules associated with embodiments of the present disclosure. The communication port(s) (960) may be any of an RS-232 port for use with a modem-based dialup connection, a 10/100 Ethernet port, a Gigabit or 10 Gigabit port using copper or fibre, a serial port, a parallel port, or other existing or future ports. The communication port(s) (960) may be chosen depending on a network, such as a Local Area Network (LAN), Wide Area Network (WAN), or any network to which the computer system (900) connects. The main memory (930) may be random access memory (RAM), or any other dynamic storage device commonly known in the art. The read-only memory (940) may be any static storage device(s), including, but not limited to, a Programmable Read Only Memory (PROM) chips for storing static information e.g., start-up or basic input/output system (BIOS) instructions for the processor (970). The mass storage device (950) may be any current or future mass storage solution, which may be used to store information and/or instructions.
[072] The bus (920) communicatively couples the processor (970) with the other memory, storage, and communication blocks. The bus (920) can be, e.g., a Peripheral Component Interconnect (PCI) / PCI Extended (PCI-X) bus, Small Computer System Interface (SCSI), universal serial bus (USB), or the like, for connecting expansion cards, drives, and other subsystems as well as other buses, such a front side bus (FSB), which connects the processor (970) to the computer system (900).
[073] Optionally, operator and administrative interfaces, e.g. a display, keyboard, and a cursor control device, may also be coupled to the bus (920) to support direct operator interaction with the computer system (900). Other operator and administrative interfaces may be provided through network connections connected through the communication port(s) (960). In no way should the aforementioned exemplary computer system (900) limit the scope of the present disclosure.
[074] Thus, the present disclosure discloses the system (102) and method (800) for automated pipe counting and inventory management. The disclosed system and method enable accurate detection, counting, and verification of pipes, including nested configurations, within a predefined storage area.
[075] While the foregoing describes various embodiments of the disclosure, other and further embodiments of the invention may be devised without departing from the basic scope thereof. The scope of the disclosure is determined by the claims that follow. The disclosure is not limited to the described embodiments, versions, or examples, which are included to enable a person having ordinary skill in the art to make and use the disclosure when combined with information and knowledge available to the person having ordinary skill in the art.

ADVANTAGES OF THE PRESENT DISCLOSURE
[076] The present disclosure significantly reduces the need for manual labour in inventory management by approximately 70% compared to traditional methods.
[077] The present disclosure increases inventory accuracy from approximately 91.3% to 97.8% in industrial environments.
[078] The present disclosure accurately handles nested pipe configurations with up to four layers, offering superior performance over previous techniques.
[079] The present disclosure enhances inventory security by enabling cross-verification between rack inventory and shipment records to prevent unauthorised removal.
[080] The present disclosure reduces dependency on network infrastructure by performing local image processing through edge computing, allowing use in connectivity-challenged environments.
[081] The present disclosure facilitates comprehensive inventory categorisation by enabling measurement of both pipe diameter and wall thickness.
, Claims:1. A pipe counting system (102) for inventory management, comprising:
at least one image acquisition unit (104) configured to acquire images of a pre-defined pipe storage area; and
a processing unit (106) operatively coupled to the image acquisition unit (104) and a memory (108), the memory storing instructions executable by the processing unit (104), wherein the processing unit (104) is configured to:
receive the images acquired by the at least one image acquisition unit (104);
apply computer vision techniques to the received images;
identify individual pipes within the images, including nested pipes;
evaluate one or more geometric properties of the identified pipes, wherein the one or more geometric properties comprise diameter and wall thickness, through contour analysis and pixel-to-millimeter calibration;
identify inner contours and outer contours of the nested pipes based on the contour analysis;
generate an inventory dataset based on count of the identified pipes in the pre-defined pipe storage area, wherein the inventory dataset comprises an association between the count and the pre-defined pipe storage area;
compare the generated inventory dataset with inventory records stored in the memory to detect unauthorised removal of the pipes from the pre-defined pipe storage area; and
generate output including current inventory status, shipment verification results, and detected discrepancies based on comparison, wherein the generated output being displayed to a computing device (110).
2. The pipe counting system (102) as claimed in claim 1, wherein the processing unit (106) is configured to apply contrast adjustment, histogram equalisation, and convolution-based edge operations to the images during application of the computer vision techniques.
3. The pipe counting system (102) as claimed in claim 1, wherein the processing unit (106) is configured to perform the pixel-to-millimeter calibration using one or more reference objects captured within the images to determine the diameter and the wall thickness of the identified pipes.
4. The pipe counting system (102) as claimed in claim 1, wherein the processing unit (106) is configured to detect and identify the individual pipes, including nested pipes, by applying a YOLOv11-based deep learning model to the preprocessed images.
5. The pipe counting system (102) as claimed in claim 1, wherein the processing unit (106) is configured to apply the contour analysis to the detected pipes to differentiate between the inner contours and the outer contours of the nested pipes for determining geometric properties.
6. A method (800) for counting pipes for inventory management, the method comprising:
acquiring (802), by at least one image acquisition unit, images of a pre-defined pipe storage area containing a plurality of pipes;
receiving (804), by a processing unit, the images acquired by the at least one image acquisition unit;
applying (806), by the processing unit, computer vision techniques to the received images;
detecting and identifying (808), by the processing unit, individual pipes within the images, including nested pipes;
evaluating (810), by the processing unit, one or more geometric properties of the identified pipes, the one or more geometric properties comprising diameter and wall thickness, by performing contour analysis and pixel-to-millimeter calibration based on reference objects present in the images;
identifying (812), by the processing unit, inner contours and outer contours of the nested pipes based on the contour analysis to differentiate between inner and outer pipe edges;
generating (814), by the processing unit, an inventory dataset based on a count of the identified pipes in the pre-defined pipe storage area, wherein the inventory dataset comprises an association between the count and the pre-defined pipe storage area;
comparing (816), by the processing unit, the generated inventory dataset with inventory records stored in a memory to detect unauthorised removal of the pipes from the pre-defined pipe storage area; and
generating (818), by the processing unit, output comprising current inventory status, shipment verification results, and detected discrepancies based on the comparison, wherein the generated output being displayed on a computing device.
7. The method (800) as claimed in claim 6, wherein applying the computer vision techniques comprises performing contrast adjustment, histogram equalisation, and convolution-based edge detection prior to detecting the pipes.
8. The method (800) as claimed in claim 6, wherein performing the pixel-to-millimeter calibration comprises using one or more reference objects captured within the images to determine the diameter and the wall thickness of the identified pipes.
9. The method (800) as claimed in claim 6, wherein detecting and identifying the individual pipes, comprising nested pipes, is performed by applying the YOLOv11-based deep learning model to the preprocessed images.
10. The method (800) as claimed in claim 6, wherein applying the contour analysis to the detected pipes comprises differentiating between the inner contours and the outer contours of the nested pipes for determining geometric properties.