System and method for automatic defect inspection and operator assessment in a circular knitting machine

Abstract

The present invention discloses a system and method for automatic defect inspection and operator assessment in a circular knitting machine (104). The system comprises at least one an external camera (102) to capture at least one image of surroundings of a circular knitting machine (104). Further, the system may comprise an internal inspection unit (106) to analyze at least one captured image to detect defects in the fabric and / or data collection unit attached to knitting machine. Furthermore, the system may comprise an automatic defect processing unit (110) to receive the at least one analyzed image data from the internal inspection unit (106) for decision making Additionally, the automatic defect processing unit (110) classifies the detected defects into preventable defects and non-preventable defects. Thereafter, the automatic defect processing unit (110) based on the classification determines whether to stop the circular knitting machine (104) or to trigger an alert for operator intervention.

Description

System and method for automatic defect inspection and operator assessment in a circular knitting machine

[0001] The field of invention generally relates to automatic defect inspection and operator assessment in a circular knitting machine. More specifically, it relates to a system and method providing real-time, edge-deployed fabric defect inspection, and operator activity recognition in the circular knitting machine.

[0002] Circular knitting machines are widely used in a textile industry for producing knitted fabrics, operate continuously and at high speeds, making real-time quality inspection essential. The circular knitting machines often encounter defects during knitting process due to various issues such as yarn breakage, needle faults, or operator errors. Typically, fabric defects are detected manually after production of entire roll is completed, which leads to significant wastage if defect levels are found to be unacceptable.

[0003] Automated inspection systems have been introduced, which use cameras and sensors to detect defects in real-time during the knitting process. However, existing systems are limited in their effectiveness, as they struggle to differentiate between preventable and non-preventable defects. This results in frequent, unnecessary stoppages, affecting machine productivity and operator efficiency.

[0004] However, current systems fail to account for operator interventions, including yarn replacement, needle changes, dial adjustments, and mechanical tuning, which are known to temporarily destabilize fabric output. As a result, defects occurring immediately after legitimate operator actions are incorrectly treated as production defects, causing unnecessary stoppages, productivity loss, and operator dissatisfaction.

[0005] Currently, existing systems do not succeed in accurately classifying defects and assessing operator’s skill level. These systems often require manual input from the operator to identify start and end of fabric rolls, and rely heavily on operator vigilance, leading to inconsistencies in defect detection and reporting. Additionally, current solutions do not adequately address the impact of operator activities on defect occurrence, making it difficult to determine the true cause of defects.

[0006] Other existing systems have tried to address this problem. However, their scope was limited to basic defect detection using infrared scanners or fixed internal cameras mounted on the rotating structure of the circular knitting machine. These approaches are hindered by issues such as high costs, maintenance challenges, and reduced reliability due to environmental contamination. Moreover, these systems do not provide insights into the operator performance or distinguish between defects caused by machine issues versus operator errors.

[0007] Existing systems also lack mechanisms to determine causality between operator actions and fabric anomalies, dynamically adjust defect tolerance based on operator activity, distinguish transient stabilization defects from persistent production defects.

[0008] The ability to classify defects into preventable and non-preventable categories offers significant advantages. By analyzing images from the fixed internal camera system, it is insufficient to determine whether a defect is preventable or non-preventable. The circular knitting machine can be stopped only when necessary, avoiding unwarranted interruptions. Preventing unnecessary stoppages reduces production losses and minimizes the risk of frustrating users to the point where they might disable the system entirely.

[0009] Thus, in light of the above discussion, it is implied that there is a need for a system and method for enhanced defect detection, classification, and operator assessment, which is reliable and does not suffer from the problems discussed above.Object of Invention

[0010] The principal object of this invention is to provide an automatic defect inspection and operator assessment in a circular knitting machine in real-time.

[0011] A further object of the invention is to provide a system and method for classifying defects as preventable or non-preventable, allowing for better decision-making on machine stoppage.

[0012] Another object of the invention is to integrate internal fabric inspection data, external operator activity monitoring, and machine sensor data.

[0013] Another object of the invention is to use temporal correlation logic to classify detected defects as either operator-conditioned transient defects, or productive defects requiring machine stoppage.

[0014] Another object of the invention is to ensure the circular knitting machine is stopped when necessary, thereby reducing unnecessary interruptions, and initiating a post-intervention defect tolerance window, upon detecting an operator intervention, during which detected fabric anomalies are suppressed, downgraded, or logged without stopping the machine.

[0015] Another object of the invention is to provide an integrated inspection system that comprises both external cameras and fabric monitoring sensors for the defect detection.

[0016] Another object of the invention is to enable real-time monitoring of the operator activity, correlating it with defect occurrences to classify machine stops effectively.

[0017] Another object of the invention is to provide a user-friendly interface for operators to interact with the system, view real-time alerts, and manage defect data.

[0018] Another object of the invention is to offer an improved reporting system that combines defect data with operator actions for accurate fabric quality assessment.

[0019] Another object of the invention is to integrate machine learning algorithms to enhance the accuracy of defect detection and classification over time.

[0020] Another object of the invention is to reduce operator dependency, ensuring consistent fabric quality regardless of the operator's skill level.

[0021] Another object of the invention is to prevent false alarms during the knitting process, allowing for smoother operations and reduced downtime.

[0022] Another object of the invention is to ensure high accuracy in defect detection, even under varying lighting conditions and fabric types.

[0023] This invention is illustrated in the accompanying drawings, throughout which, like reference letters indicate corresponding parts in the various figures.

[0024] The embodiments herein will be better understood from the following description with reference to the drawings, in which:Fig. 1

[0025] depicts / illustrates a system for an automatic defect inspection and an operator assessment in the circular knitting machine, in accordance with an embodiment of the present disclosure;Fig. 2

[0026] depicts / illustrates a block diagram of an automatic defect processing unit, in accordance with an embodiment of the present disclosure;Fig. 3

[0027] depicts / illustrates placement of an external camera to monitor surrounding of a circular knitting machine, in accordance with an embodiment of the present disclosure; and;Fig. 4

[0028] illustrates a method for the automatic defect inspection and the operator assessment in the circular knitting machine, in accordance with an embodiment of the present disclosure.Statement of Invention

[0029] The present invention comprises a system and method for automatic defect inspection and operator assessment in a circular knitting machine. The system comprises a circular knitting machine to produce knitted fabric. At least one external camera is attached to the circular knitting machine to capture at least one image of surroundings of the circular knitting machine. At least one internal inspection unit is used to monitor the circular knitting machine and record inspection data to detect fabric defects during production.

[0030] An automatic defect processing unit receives and processes inspection data from the internal inspection unit to detect, classify and analyse defects into preventable defects and non-preventable defects for automatic defect inspection and operator assessment. A data collection unit is used to tap machine related data from the knitting machine.Detailed Description

[0031] The embodiments herein and the various features and advantageous details thereof are explained more fully with reference to the non-limiting embodiments that are illustrated in the accompanying drawings and / or detailed in the following description. Descriptions of well-known components and processing techniques are omitted so as to not unnecessarily obscure the embodiments herein. The examples used herein are intended merely to facilitate an understanding of ways in which the embodiments herein may be practiced and to further enable those of skill in the art to practice the embodiments herein. Accordingly, the examples should not be construed as limiting the scope of the embodiments herein.

[0032] The present invention discloses a real-time, edge-deployed system and method for fabric defect inspection, operator activity recognition, and causality-based machine control in circular knitting machines.

[0033] The invention discloses a causality-aware defect inspection and machine control system for circular knitting machines.

[0034] The system integrates internal fabric inspection data, external operator activity monitoring, machine sensor data, and applies temporal correlation logic to classify detected defects as either operator-conditioned transient defects, or productive defects requiring machine stoppage.

[0035] Upon detecting an operator intervention, the system initiates a post-intervention defect tolerance window, during which detected fabric anomalies are suppressed, downgraded, or logged without stopping the machine. Once the tolerance window expires, or if defects persist beyond defined thresholds, the system automatically stops the machine.

[0036] The invention is implemented primarily on edge computing hardware, producing a direct technical effect on machine control, with optional cloud-based learning feedback.

[0037] The system uses an external camera to monitor surroundings of the circular knitting machine area which may be combined with an automatic inspection system that detects defects in real-time and a data collection unit to tap machine related data from knitting machine. The internal camera is placed on the circular knitting machine to detect the defects in the fabric. The defects are then classified into preventable defects and non-preventable defects. By classifying these defect types, the system can automatically stop the circular knitting machine for preventable defects, reducing fabric waste, while alerting operators for non-preventable defects, improving efficiency. The present invention enhances defect reporting and decision-making, minimizing unnecessary machine stoppages and providing a clear assessment of the inspection system’s impact. The present system ensures the circular knitting machine is stopped only when necessary, reducing unnecessary interruptions, thereby avoiding unnecessary stoppages that could lead to production losses and potential user frustration, which might result in the system being disabled.

[0038] depicts / illustrates a system 100 for an automatic defect inspection and an operator assessment in the circular knitting machine 104, in accordance with an embodiment of the present disclosure;

[0039] The system 100 comprises at least one external camera 102, at least one circular knitting machine 104, at least one internal inspection unit 106, a communication network 108, an automatic defect processing unit and a data collection unit 110 connected to knitting machine.

[0040] The at least one external camera 102 may be an external camera 102 / 1, an external camera 102 / 2, …and external camera 102 / n, as per user requirements.

[0041] The at least one external camera 102 is configured to capture at least one image of surroundings of the circular knitting machine 104 to monitor operator presence and activity.

[0042] In an embodiment, the at least one external camera 102 may ensure coverage of the circular knitting machine’s 104 vicinity to detect when the operator accesses the circular knitting machine 104.

[0043] The at least one external camera 102 may evaluate the operator's skill level and attentiveness within their work environment using AI-based activity recognition, and detect and timestamp operator interventions.

[0044] The circular knitting machine 104 produces knitted fabric and operates in a cylindrical fashion to create tube-shaped fabric that is flattened and wound onto a cylindrical rod. The circular knitting machine's performance can directly influence types of defects in the knitted fabric.

[0045] In an embodiment, the circular knitting machine 104 comprises at least one light (not shown), at least one fixed structure (not shown), a knitted fabric (not shown), and at least one rotating structure (not shown).

[0046] The at least one light is configured to illuminate the fabric uniformly to ensure clear visibility for the at least one internal camera and the at least one external camera 102. The at least one light may be mounted on a rotating or static structure of the circular knitting machine 104 parallel to the fabric.

[0047] The circular knitting machine 104 is monitored by the at least one internal inspection unit 106 for detecting fabric defects during production.

[0048] The internal inspection unit 106 is configured to receive and analyze the at least one captured image to detect defects in the fabric for decision making.

[0049] The internal inspection unit 106 may work in conjunction with the external camera 102 to detect and classify the fabric defects. The internal inspection unit 106 provides detailed inspection data that is then processed by the automatic defect processing unit 110 to decide whether the defects are preventable or non-preventable.

[0050] The at least one internal inspection unit 106 comprises at least one internal fabric-facing camera and at least one sensor.

[0051] The at least one internal fabric-facing camera may be mounted on the at least one fixed structure or the rotating structure of the circular knitting machine 104.

[0052] The at least one internal camera is configured to capture high-resolution images of the knitted fabric surface to detect surface-level defects, monitor the fabric surface for visible defects like holes, snags, yarn tension issues, misfeeds, or misalignments, utilize advanced techniques, such as infrared imaging or ultra-violet (UV) light sensors, to detect defects not visible under normal lighting conditions, such as hidden yarn breaks or irregularities in the knit pattern, monitor fabric continuity and detect anomalies in real-time, ensuring early identification of defects.

[0053] The at least one internal camera is configured to capture at least one image of a fabric being produced by the circular knitting machine 104.

[0054] In an embodiment, the at least one internal camera may capture the at least one image of the knitted fabric to identify visible defects such as yarn breaks, needle lines, and holes.

[0055] In an embodiment, the at least one internal camera may capture operator's actions, such as yarn changes, needle adjustments, and fabric removal, to correlate these activities with detected defects.

[0056] The at least one internal camera comprises at least one of a color sensor, an infrared sensor, a high-resolution sensor, or a thermal sensor.

[0057] The at least one sensor is configured for recording sensor data related to physical anomalies in the fabric during the knitting process, such as excessive yarn tension, missed yarns, or skipped stitches, monitor machine parameters, such as needle positions, yarn feed rates, and knitting speed, to identify any mechanical issues that may lead to defects, measure fabric tension to ensure it is within the desired range.

[0058] The at least one sensor may be at least one of tension sensor, vibration sensor, proximity sensor and pressure sensor.

[0059] The communication network 108 facilitates data transmission between the knitting machine, the automatic defect processing unit 110, and other external systems.

[0060] The communication network 108 connects the circular knitting machine 104 and the automatic defect processing unit 110 to a local or cloud-based network through a network interface, allowing for remote monitoring and data logging.

[0061] The communication network 108, through at least one data transmission protocol, ensures that the data from at least one external camera 102 and other inspection units are sent to the automatic defect processing unit 110 for processing without delay.

[0062] In an embodiment, the communication network 108 may include wired and wireless communication, including but not limited to, GPS, GSM, LAN, Wi-fi compatibility, Bluetooth low energy as well as NFC. The wireless communication may further comprise one or more of Bluetooth (registered trademark), ZigBee (registered trademark), a short-range wireless communication such as UWB, a medium-range wireless communication such as WiFi (registered trademark) or a long-range wireless communication such as 3G / 4G or WiMAX (registered trademark), according to the usage environment.

[0063] In an embodiment, the automatic defect processing unit 110 is configured to receive the at least one analyzed image or inspection data from the internal inspection unit 106. Additionally, the automatic defect processing unit 110 classifies the detected defects into preventable defects and non-preventable defects. Thereafter, the automatic defect processing unit 110 based on the classification determines whether to stop the circular knitting machine 104 or to trigger an alert for operator intervention. Historical operator behavior and defect outcomes are used to update defect classification parameters

[0064] Further, upon detection of operator activity, the automatic defect processing unit 110 is configured to: initiate a post-intervention tolerance window, dynamically modify defect classification thresholds, and conditionally suppress or downgrade defects detected within the window. In case defects persist beyond the tolerance window, the defects are reclassified as productive defects.

[0065] Defect tolerance duration and classification thresholds are configurable by an operator. The automatic defect processing unit 110 assigns a confidence score to each defect classification and controls machine stoppage based on the confidence score.

[0066] depicts / illustrates a block diagram of an automatic defect processing unit 110, in accordance with an embodiment of the present disclosure.

[0067] The automatic defect processing unit 110 comprises a data acquisition module 202, a signal processing module 204, an image processing module 206, an Artificial Intelligence (AI) processing module 208, a defect classification module 210, a controller interface module 212, a user interface module 214, a communication module 216, and a memory module 218.

[0068] The data acquisition module 202 is configured to collect real-time data from the at least one sensor. Further, the data acquisition module 202 may continuously monitor the fabric production, capturing the real-time data essential for defect detection and classification.

[0069] The signal processing module 204 is configured to process raw sensor signals received from the at least one sensor, convert analog signals to digital form, filter out noise, and calibrate the sensor data for accurate interpretation.

[0070] The image processing module 206 comprises at least one of: image enhancement algorithm, defect detection filters, and an edge detection algorithm.

[0071] The image processing module 206 is configured to analyze images captured by the at least one external camera 102 and the at least one internal camera.

[0072] Further, the image processing module 206 is configured to detect at least one visible defect and enhance image features for better defect identification.

[0073] The at least one visible defect may be holes, misfeeds, and yarn breaks.

[0074] Furthermore, the image processing module 206 comprises edge detection and pattern recognition algorithms to identify irregularities on the fabric surface. The detection and pattern recognition algorithms may comprise at least one of Convolutional Neural Networks (CNNs), Hough Transform, Scale-Invariant Feature Transform (SIFT), Histogram of Oriented Gradients (HOG), K-Means Clustering, Principal Component Analysis (PCA).

[0075] The Artificial Intelligence (AI) processing module 208 is configured to analyze the data from the image processing module 206 and the signal processing module 204 using AI algorithms to predict and classify fabric defects.

[0076] The Artificial Intelligence (AI) processing module 208 comprises at least one of neural network-based defect recognition model, machine learning classifiers, and a predictive analysis engine, and an artificial intelligence model trained to recognize human actions from video data operator activity detection. The machine learning classifiers may be at least one of Support Vector Machines (SVM) or Convolutional Neural Networks (CNN).

[0077] Further, the Artificial Intelligence (AI) processing module 208 is configured to continuously learn from new data from the image processing module 206 and the signal processing module 204 to improve defect detection accuracy over time, adapting to variations in fabric types and production conditions.

[0078] The defect classification module 210 comprises a defect classification engine and a rules-based decision model.

[0079] The defect classification module 210 is configured to classify detected defects into preventable or non-preventable defects based on historical data, machine conditions, and operator actions.

[0080] The preventable defects are defects which typically occur due to standard knitting machine operations or conditions that do not require immediate operator intervention. Preventable defects are inherent to the knitting process and are generally expected or tolerated within certain limits.

[0081] The preventable defects may comprise at least one of yarn break, lycra break, needle lines at start, or count mix-ups at start.

[0082] The non-preventable defects comprise defects which result from human error, improper machine settings, or unexpected machine behavior. The non-preventable defects require operator intervention to diagnose the issue, correct the problem, and restart the machine.

[0083] The non-preventable defects may comprise at least one of holes in fabric, bent needle marks, lycra jump, or manual yarn change issues.

[0084] The defect classification module 210 is configured to utilize the defect classification engine and the rules-based decision model to accurately label defects and determine necessary corrective actions.

[0085] The defect classification module 210 provides improved decision making, enhanced efficiency and better-quality control.

[0086] The controller interface module 212 is configured to communicate with the circular knitting machines 104 control system to execute commands such as stopping the machine, adjusting settings, or triggering alarms. Further, override manual controls if a critical defect is detected, ensuring immediate stoppage of the machine to prevent further fabric damage.

[0087] The user interface module 214 is configured to provide an interface comprising at least one of a touchscreen display, control buttons, and a visual alert module.

[0088] The user interface module 214 is configured to provide real-time feedback to the operator, displaying detected defects, machine status, and recommended corrective actions.

[0089] Further, the user interface module 214 is configured to enable the operator to input data, such as roll start / end points or adjustments to machine settings, and to acknowledge alarms or alerts.

[0090] The communication module 216 comprises may comprise wireless communication module 216 such as Wi-Fi or Bluetooth and wired communication module 216 such as Ethernet or USB.

[0091] The communication module 216 is configured to enable data exchange between the internal inspection unit 106, the at least one external camera 102, the automatic defect processing unit 110, and remote monitoring systems.

[0092] The communication module 216 is also configured to transmit inspection reports, defect logs, and real-time machine data to a centralized server for further analysis and quality control.

[0093] The memory module 218 is configured to store captured images, sensor data, inspection logs, and defect classification results for historical analysis and reporting.

[0094] depicts / illustrates placement of an external camera 102 to monitor surrounding of a circular knitting machine 104, in accordance with an embodiment of the present disclosure;

[0095] The at least one external camera 102a, 102b is positioned at an optimal height and angle to capture real-time images of the surroundings of the circular knitting machine, while also observing the operator’s workspace. This strategic placement of at least one external camera 102 allows the system to detect visible fabric defects like yarn breaks and needle lines, and simultaneously track operator activities such as yarn changes or machine adjustments. By correlating these observations, the system can distinguish between defects inherent to the knitting process (preventable defects) and those caused by operator actions (non-preventable defects), enhancing defect analysis, reducing unnecessary machine stoppages, and minimizing fabric waste.

[0096] illustrates a method for the automatic defect inspection and the operator assessment in the circular knitting machine 104, in accordance with an embodiment of the present disclosure.

[0097] The method begins with capturing, by at least one an external camera 102, images of surroundings of a circular knitting machine 104, as depicted at step 402. Subsequently, the method 400 discloses analyzing the captured images using an internal inspection unit 106 to detect defects in the fabric, as depicted at step 404. The method discloses monitoring operator presence and operator activity using at least one external camera.

[0098] Additionally, the method 400 discloses receiving, by an automatic defect processing unit 110, the analyzed image data from the internal inspection unit 106, as depicted at step 406. The method discloses detecting operator interventions based on monitored operator activity.

[0099] Furthermore, the method 400 discloses classifying, by the automatic defect processing unit 110, the detected defects into preventable defects and non-preventable defects, as depicted at step 408. The method discloses temporally correlating detected operator interventions with detected fabric defects. The method discloses classifying fabric defects based on the temporal correlation.

[0100] The method discloses selectively controlling operation of the circular knitting machine based on the classification.

[0101] Thereafter, the method 400 discloses determining, by the automatic defect processing unit 110, based on the classification to stop the circular knitting machine 104 or to trigger an alert for operator intervention, as depicted at step 410.

[0102] The advantages of the current invention include enhanced real-time detection of fabric defects, thereby reducing wastage by identifying issues at an early stage.

[0103] An additional advantage is that the system differentiates between preventable and non-preventable defects, thereby minimizing unnecessary stoppages.

[0104] An additional advantage is that the system offers precise classification of the defects, thereby enabling more accurate reporting and better-quality control.

[0105] An additional advantage is the automated analysis of operator skill and behavior, which aids in assessing and improving operator performance.

[0106] An additional advantage is the integration with existing inspection systems, thereby allowing for enhanced decision-making based on correlated data.

[0107] An additional advantage is that the system reduces operator workload by providing automated alerts and suggestions for corrective actions.

[0108] An additional advantage is its capability to generate detailed defect reports that assist in quality audits and process improvements.

[0109] Applications of the current invention include fabric inspection in textile manufacturing, operator performance monitoring, quality assurance in knitting mills, predictive maintenance for knitting machines, and process optimization in circular knitting operations.

[0110] The foregoing description of the specific embodiments will so fully reveal the general nature of the embodiments herein that others can, by applying current knowledge, readily modify and / or adapt for various applications such specific embodiments without departing from the generic concept, and, therefore, such adaptations and modifications should and are intended to be comprehended within the meaning and range of equivalents of the disclosed embodiments. It is to be understood that the phraseology or terminology employed herein is for the purpose of description and not of limitation. Therefore, while the embodiments herein have been described in terms of preferred embodiments, those skilled in the art will recognize that the embodiments herein can be practiced with modification within the scope of the embodiments as described here.

Claims

A system for automatic defect inspection and operator assessment in a circular knitting machine, comprising:a circular knitting machine (104) to produce knitted fabric;at least one external camera (102) attached to the circular knitting machine (104) to capture at least one image of surroundings of the circular knitting machine (104) to monitor operator presence and operator activity relative to the circular knitting machine;at least one internal inspection unit (106) to monitor the circular knitting machine (104) and record inspection data to detect fabric defects during production; andan automatic defect processing unit (110) operatively coupled to the internal inspection unit (106) and the external camera (102), the automatic defect processing unit being configured to:receive and process the image from the external camera to detect operator interventions;receive and process inspection data from the internal inspection unit (106) to detect, classify and analyse defects;temporally correlate detected operator interventions with detected fabric defects;dynamically classify detected fabric defects based on the temporal correlation into preventable defects and non-preventable defects for automatic defect inspection and operator assessment; andselectively control operation of the circular knitting machine based on the defect classification.The fabric quality control system as claimed in claim 1, wherein the automatic defect processing unit initiates a post-intervention defect tolerance window upon detection of operator intervention, during which detected fabric defects are prevented from triggering machine stoppage,wherein fabric defects persisting beyond the tolerance window are reclassified as productive defects and trigger automatic machine stoppage,wherein the operator activity comprises at least one of: yarn replacement, needle adjustment, dial modification, and mechanical tuning,wherein defect tolerance duration and classification thresholds are configurable by an operator,wherein the automatic defect processing unit (110) assigns a confidence score to each defect classification and controls machine stoppage based on the confidence score, andwherein historical operator behavior and defect outcomes are used to update defect classification parameters.The fabric quality control system as claimed in claim 1, comprising:a data collection unit to tap machine related data from knitting machine;at least one light mounted on a rotating or static structure of the circular knitting machine (104) parallel to the fabric, wherein the light is configured to illuminate the fabric uniformly to ensure clear visibility for the at least one internal camera and the at least one external camera (102);a communication network (108) facilitates data transmission between the knitting machine, the automatic defect processing unit (110), and other external systems.The fabric quality control system as claimed in claim 2,at least one internal camera disposed in the internal inspection unit (106) to capture at least one of high-resolution images of the knitted fabric surface and operator's actions, comprising yarn changes, needle adjustments, and fabric removal, to correlate the activities with detected defects,wherein the internal camera comprises at least one of a color sensor, an infrared sensor, a high-resolution sensor, a thermal sensor, and ultra-violet (UV) light sensors;at least one sensor disposed in the internal inspection unit (106) to record sensor data related to physical anomalies in the fabric during the knitting process, comprising excessive yarn tension, missed yarns, skipped stitches, monitor machine parameters comprising needle positions, yarn feed rates, and knitting speed, in order to identify any mechanical issues that lead to defects,wherein the sensor comprises at least one of: tension sensor, vibration sensor, proximity sensor and pressure sensor;the at least one external camera (102a), (102b) positioned at an optimal height and angle to capture real-time images of the surroundings of the circular knitting machine, and the operator’s workspace, to detect visible fabric defects like yarn breaks and needle lines, and simultaneously track operator activities comprising yarn changes or machine adjustments.The fabric quality control system as claimed in claim 1, wherein the automatic defect processing unit (110) comprises at least one of:a data acquisition module (202) to collect sensor data from the sensor;a signal processing module (204) configured to process raw sensor data, convert analog signals to digital form, filter out noise, and calibrate the sensor data;an image processing module (206) to receive and analyze images captured by the external camera (102) and the internal camera to detect at least one visible defect and enhance image features for better defect identification,an Artificial Intelligence (AI) processing module 208 to analyze the data from the image processing module (206) and the signal processing module (204) using AI algorithms to predict and classify fabric defects,wherein the AI processing module 208 comprises at least one of:a neural network-based defect recognition model;machine learning classifiers comprising at least one of Support Vector Machines (SVM) or Convolutional Neural Networks (CNN);a predictive analysis engine to continuously learn from new data from the image processing module (206); andan artificial intelligence model trained to recognize human actions from video data operator activity detection; andthe signal processing module (204) to improve defect detection accuracy over time, adapting to variations in fabric types and production conditions;a defect classification module 210 comprising:a defect classification engine to classify detected defects into preventable or non-preventable defects based on historical data, machine conditions, and operator actions; anda rules-based decision model to determine necessary corrective actions,a controller interface module 212 to communicate with the circular knitting machines (104) control system to execute commands such as stopping the machine, adjusting settings, or triggering alarms;a user interface module 214 to provide an interface comprising at least one of a touchscreen display, control buttons, and a visual alert module, to provide real-time feedback to the operator, displaying detected defects, machine status, and recommended corrective actions and enable the operator to input data comprising roll start / end points or adjustments to machine settings;a communication module 216 to enable data exchange between the internal inspection unit (106), the at least one external camera (102), the automatic defect processing unit (110), and remote monitoring systems; anda memory module 218 to store captured images, sensor data, inspection logs, and defect classification results for historical analysis and reporting:The fabric quality control system as claimed in claim 4, wherein the image processing module (206) comprises at least one of:an image enhancement algorithm to enhance image features for better defect identification;edge detection and pattern recognition algorithms to identify irregularities on the fabric surface, wherein the detection and pattern recognition algorithms may comprise at least one of Convolutional Neural Networks (CNNs), Hough Transform, Scale-Invariant Feature Transform (SIFT), Histogram of Oriented Gradients (HOG), K-Means Clustering, Principal Component Analysis (PCA),wherein the visible defect comprises at least one of holes, misfeeds, and yarn breaks;wherein the preventable defects comprise defects inherent to the knitting process which are expected or tolerated within certain limit comprising at least one of yarn break, lycra break, needle lines at start, or count mix-ups at start,wherein the non-preventable defects comprise defects which result from human error, improper machine settings, or unexpected machine behavior comprising at least one of holes in fabric, bent needle marks, lycra jump, or manual yarn change issues.A method for automatic defect inspection and operator assessment in a circular knitting machine a circular knitting machine, comprising:producing knitted fabric using a circular knitting machine (104);attaching at least one external camera (102) to the circular knitting machine (104) to capture at least one image of surroundings of the circular knitting machine (104);using at least one internal inspection unit (106) to monitor the circular knitting machine (104); recording inspection data to detect fabric defects during production;receiving and processing inspection data from the internal inspection unit (106); anddetecting and classifying defects into preventable defects and non-preventable defects.The method for automatic defect inspection and operator assessment as claimed in claim 6, comprising:mounting at least one light on a rotating or static structure of the circular knitting machine (104) parallel to the fabric;illuminating the fabric uniformly to ensure clear visibility for the at least one internal camera and the at least one external camera (102);a communication network (108) 10 facilitates data transmission between the knitting machine, the automatic defect processing unit (110), and other external systems.The method for automatic defect inspection and operator assessment as claimed in claim 7, comprising:capturing at least one of high-resolution images of the knitted fabric surface and operator's actions by using at least one internal camera, comprising yarn changes, needle adjustments, and fabric removal, to correlate the activities with detected defects, wherein the internal camera comprises at least one of a color sensor, an infrared sensor, a high-resolution sensor, a thermal sensor, and ultra-violet (UV) light sensors;recording sensor data related to physical anomalies in the fabric during the knitting process by using at least one sensor disposed, comprising excessive yarn tension, missed yarns, skipped stitches, monitor machine parameters comprising needle positions, yarn feed rates, and knitting speed, in order to identify any mechanical issues that lead to defects, wherein the sensor comprises at least one of: tension sensor, vibration sensor, proximity sensor and pressure sensor;positioning the at least one external camera (102a), (102b) at an optimal height and angle to capture real-time images of the surroundings of the circular knitting machine, and the operator’s workspace; anddetecting visible fabric defects like yarn breaks and needle lines, and simultaneously tracking operator activities comprising yarn changes or machine adjustments.The method for automatic defect inspection and operator assessment as claimed in claim 6, comprising:collecting sensor data from the sensor;processing raw sensor data;converting analog signals to digital form;filtering out noise;calibrating the sensor data;receiving and analyzing images captured by the external camera (102) and the internal camera;detecting at least one visible defect and enhancing image features for better defect identification,analyzing the data from the image processing module (206) and the signal processing module (204) using AI algorithms to predict and classify fabric defects, wherein the AI processing module 208 comprises at least one of: a neural network-based defect recognition model, machine learning classifiers comprising at least one of Support Vector Machines (SVM) or Convolutional Neural Networks (CNN), and a predictive analysis engine to continuously learn from new data from the image processing module (206) and the signal processing module (204) to improve defect detection accuracy over time, adapting to variations in fabric types and production conditions;using a defect classification engine to classify detected defects into preventable or non-preventable defects based on historical data, machine conditions, and operator actions; andusing a rules-based decision model to determine necessary corrective actions,communicating with the circular knitting machines (104) control system to execute commands such as stopping the machine, adjusting settings, or triggering alarms;providing an interface comprising at least one of a touchscreen display, control buttons, and a visual alert module, to provide real-time feedback to the operator, displaying detected defects, machine status, and recommended corrective actions and enable the operator to input data comprising roll start / end points or adjustments to machine settings;enabling data exchange between the internal inspection unit (106), the at least one external camera (102), the automatic defect processing unit (110), and remote monitoring systems; andstoring captured images, sensor data, inspection logs, and defect classification results for historical analysis and reporting:The method for automatic defect inspection and operator assessment as claimed in claim 6, comprising:using an image enhancement algorithm to enhance image features for better defect identification;using edge detection and pattern recognition algorithms to identify irregularities on the fabric surface, wherein the detection and pattern recognition algorithms may comprise at least one of Convolutional Neural Networks (CNNs), Hough Transform, Scale-Invariant Feature Transform (SIFT), Histogram of Oriented Gradients (HOG), K-Means Clustering, Principal Component Analysis (PCA),wherein the visible defect comprises at least one of holes, misfeeds, and yarn breaks;wherein the preventable defects comprise defects inherent to the knitting process which are expected or tolerated within certain limits comprising at least one of yarn break, lycra break, needle lines at start, or count mix-ups at start,wherein the non-preventable defects comprise defects which result from human error, improper machine settings, or unexpected machine behavior comprising at least one of holes in fabric, bent needle marks, lycra jump, or manual yarn change issues.

Images