Material processing

Abstract

A method for controlling equipment used in processing material, the method including, in one or more processing devices: acquiring image data from an image sensor, the image data being indicative of one or more images of processed material output from the equipment; analyzing the image data to determine at least one characteristic of the processed material; and controlling at least one operating parameter of the equipment based on the at least one characteristic; the method including, in one or more processing devices: analyzing the at least one characteristic; and controlling the at least one operating parameter using results of, the analysis; the method including, in one or more processing devices: quantifying the at least one characteristic to determine a characteristic value; and, controlling the at least one operating parameter using the characteristic value.

Description

MATERIAL PROCESSINGBackground of the Invention

[0001] The present invention relates to a method and apparatus for processing material, and in one particular example, to a method and apparatus for recycling materials such as plastics.Description of the Prior Art

[0002] The reference in this specification to any prior publication (or information derived from it), or to any matter which is known, is not, and should not be taken as an acknowledgement or admission or any form of suggestion that the prior publication (or information derived from it) or known matter forms part of the common general knowledge in the field of endeavour to which this specification relates.

[0003] Polyvinyl chloride (PVC), renowned for its durability and versatility, with recycled PVC playing a pivotal role in various applications, such as vinyl flooring, profiles, cables, and membranes. Virgin PVC resin is primarily utilised in the fabrication of water and sewerage pipes, plumbing fixtures, electrical conduits, and irrigation systems.

[0004] Despite being widely utilised, PVC exhibits low recycling and recovery rates that are significantly inferior to that of other polymers. The intricacy of PVC recycling is attributed to its myriad additives, indispensable for processing and endowing it with distinctive properties, which in turn poses formidable challenges to recycling. As an example, there are approximately 5000 different formulations in widespread circulation in Australia and successful recycling mandates precise knowledge of PVC composition, rendering the process technically demanding, time-consuming, and financially burdensome.

[0005] Because of these challenges, PVC compounders and manufacturers with recycling capacity prefer to accept pre-granulated PVC, as there is no way to know the composition and product origin of material that has already been granulated. This creates a secondary problem of transport logistics, where recovering and processing post-consumer PVC waste such as pipe or profile waste becomes cost-prohibitive, as it is relatively light weight and high volume.

[0006] Because of these challenges PVC compounders and manufacturers prefer to use virgin material, contributing to the low recycling rates.Summary of the Present Invention

[0007] In one broad form, an aspect of the present invention seeks to provide a method for controlling equipment used in processing material, the method including, in one or more processing devices: acquiring image data from an image sensor, the image data being indicative of one or more images of processed material output from the equipment; analysing the image data to determine at least one characteristic of the processed material; and, controlling at least one operating parameter of the equipment based on the at least one characteristic.

[0008] In one embodiment the method includes, in one or more processing devices: analysing the at least one characteristic; and, controlling the at least one operating parameter using results of the analysis.

[0009] In one embodiment the method includes, in one or more processing devices: quantifying the at least one characteristic to determine a characteristic value; and, controlling the at least one operating parameter using the characteristic value.

[0010] In one embodiment the processed material includes an extrusion and wherein the method includes, in one or more processing devices: analysing the image data to identify an extrusion edge; and, determining a characteristic value by quantifying an edge straightness.

[0011] In one embodiment the method includes, in one or more processing devices: analysing the image data to identify material surface features; and, determining a characteristic value by quantifying a density of surface features.

[0012] In one embodiment the method includes, in one or more processing devices: analysing the image data to identify material colours; and, determining a characteristic value by quantifying a colour homogeneity.

[0013] In one embodiment the method includes, in one or more processing devices, analysing the image data by executing at least one computational model, the at least one computationalmodel being trained by applying machine learning to reference images of processed material having different characteristics.

[0014] In one embodiment the method includes, in one or more processing devices: determining the at least one characteristic; adjusting the at least one operating parameter; determining a change in the least one characteristic caused by adjusting the at least one operating parameter; and, controlling the at least one operating parameter based in the change in the at least one characteristic.

[0015] In one embodiment the method includes, in one or more processing devices: quantifying the at least one characteristic to determine a characteristic value; adjusting the at least one operating parameter; quantifying a change in the least one characteristic by determining a change in characteristic value caused by adjusting the at least one operating parameter; and, controlling the at least one operating parameter using the change in characteristic value.

[0016] In one embodiment the method includes, in one or more processing devices, analysing the at least one characteristic by executing at least one computational model, the at least one computational model being trained by applying machine learning to reference characteristics derived from processed material processed using different operating parameters.

[0017] In one embodiment the at least one characteristic includes at least one of: a material shape; a material structure; a material colour; material surface features; and, surface bubbles.

[0018] In one embodiment the at least one operating parameter includes: a processing temperature; a processing pressure; a processing rate; and, an amount of at least one processing additive.

[0019] In one embodiment the method is performed using processing equipment including a material extruder configured to extrude material in accordance with the at least one operating parameter.

[0020] In one embodiment the method is performed using a material extruder including: a hopper configured to receive granular material; a screw configured to urge material through a barrel to an outlet; and, a heater configured to heat at least one of the barrel and the outlet.

[0021] In one embodiment the method includes imaging the processed material by transporting the processed material past an image sensor.

[0022] In one embodiment the method includes transporting the processed material past an image sensor by extruding the processed material onto a conveyor belt.

[0023] In one embodiment the method includes, imaging the processed material by using: a housing including an opening configured to receive processed material; an image sensor provided in the housing; and, a light source provided in the housing, the light source being configured to expose the processed material to consistent illumination.

[0024] In one embodiment the material is at least one of: a polymeric material; Polyvinyl Chloride (PVC); a material undergoing recycling; a granulated material; and, a granulated PVC material.

[0025] In one broad form, an aspect of the present invention seeks to provide a method for recycling material, the method including, in one or more processing devices: acquiring image data from an image sensor, the image data being indicative of one or more images of processed material output from the equipment; analysing the image data to determine at least one characteristic of the processed material; and, controlling at least one operating parameter of the material processing equipment based on the at least one characteristic to thereby recycle the material.

[0026] In one broad form, an aspect of the present invention seeks to provide an apparatus for controlling equipment used in processing material, the apparatus including one or more processing devices configured to: acquire image data from an image sensor, the image data being indicative of one or more images of processed material output from the equipment; analyse the image data to determine at least one characteristic of the processed material; and, control at least one operating parameter of the material processing equipment based on the at least one characteristic.

[0027] In one broad form, an aspect of the present invention seeks to provide an apparatus for processing material, the apparatus including: equipment configured to process the material using at least one operating parameter; an image sensor configured to capture images ofprocessed material output from the equipment and generate image data indicative of one or more images; one or more processing devices configured to: analyse the image data to determine at least one characteristic of the processed material; and, control at least one operating parameter of the equipment based on the at least one characteristic.

[0028] In one broad form, an aspect of the present invention seeks to provide an apparatus for recycling material, the apparatus including: equipment configured to process the material using at least one operating parameter; an image sensor configured to capture images of processed material output from the equipment and generate image data indicative of one or more images; one or more processing devices configured to: analyse the image data to determine at least one characteristic of the processed material; and, control at least one operating parameter of the equipment based on the at least one characteristic to thereby recycle the material.

[0029] It will be appreciated that the broad forms of the invention and their respective features can be used in conjunction and / or independently, and reference to separate broad forms is not intended to be limiting. Furthermore, it will be appreciated that features of the method can be performed using the system or apparatus and that features of the system or apparatus can be implemented using the method.Brief Description of the Drawings

[0030] Various examples and embodiments of the present invention will now be described with reference to the accompanying drawings, in which: -

[0031] Figure 1 is a schematic diagram of an example of an apparatus including equipment for processing material;

[0032] Figure 2 is a flow chart of an example of a method for controlling equipment used in processing material;

[0033] Figure 3 is a schematic diagram of a specific example of an apparatus including equipment for recycling PVC; and,

[0034] Figure 4 is a flow chart of a specific example of a method for controlling equipment for recycling PVC.Detailed Description of the Preferred Embodiments

[0035] An example of an apparatus including equipment for processing material will now be described with reference to Figure 1.

[0036] For the purpose of illustration, the following examples will focus on recycling of plastic materials, and in particular PVC. However, it will be appreciated that the techniques could be applied more broadly to manufacturing a wide range of materials, and whilst the techniques described herein are particularly beneficial when applied to PVC, this is not intended to be limiting.

[0037] In this example, the apparatus 100 includes processing equipment 110, which is configured to receive and process material. In one example, this will involve having the processing equipment 110 receive feedstock, such as granulated plastics material, and then processing the feedstock, for example through the application of heat and / or pressure, and optionally by combining the feedstock with additives, such as binding agents, or the like. The nature of the processing equipment will vary depending on the preferred implementation and the nature of the material being processed, but in one example, the equipment includes an extruder, as will be described in more detail below.

[0038] The apparatus 110 further includes an image sensor 120, which is configured to capture images of processed material M, output from the processing equipment. The image sensor 120 could be of any appropriate form, and could include a camera, CMOS sensor and associated optics, or the like.

[0039] The apparatus 100 also typically includes one or more processing devices 130, optionally forming part of one or more processing systems. The processing devices could form part of a controller used to control the processing equipment 110, or could alternatively be separate to, but in communication with a controller. Whilst the system can use multiple processing devices, with processing performed by one or more of the devices, for the purpose of ease of illustration, the following examples will refer to a single device, but it will be appreciated that reference to a singular processing device should be understood to encompass multiple processing devices and vice versa, with processing being distributed between the devices as appropriate.

[0040] An example of a process for controlling the equipment 110 will now be described with reference to Figure 2.

[0041] In this example, at step 200 the processing device acquires image data from the image sensor 120, with the image data being indicative of one or more images of processed material M output from the equipment 110.

[0042] At step 210, the processing device analyses the image data to determine at least one characteristic of the processed material. The manner in which this is achieved will vary depending on the preferred implementation, the nature of the material being processed and the characteristics being assessed. For example, the characteristics could include the colour of the material, in which case this will simply involve analysing the image to determine the material colour. Alternatively, the characteristics could include the shape of the material and / or the presence of surface features, such as bubbles, or other defects. In this case, the analysis may involve performing edge detection to identify edges of the material, or the presence of the surface features. The analysis can be performed using any suitable image processing technique, and could involve the use of computational models, such as neural networks or similar, which have been trained to identify and / or quantify features of interest.

[0043] At step 220, the processing device operates to control at least one operating parameter of the equipment, based on the at least one characteristic. For example, if the material includes excessive surface defects, this could indicate that the temperature being used to process the material is too high, in which case the processing device 130 can adjust the temperature being used, thereby reducing the number of surface defects.

[0044] Accordingly, the above described approach uses images of material output from the processing equipment to assess the effectiveness of the material processing. This can then be used to control the processing equipment, so that the quality of the processed material is used to provide feedback so as to optimise operation of the processing equipment. This approach allows operation of the processing equipment 110 to be dynamically adjusted in real time, so that the equipment automatically adapts to changes in properties of the feedstock material. Consequently, particularly when recycling granulated PVC material, this allows operation of the processing equipment to be controlled so that knowledge of the formulation and structureof the granulated PVC material is not required in order for the material to be processed in an optimum manner.

[0045] A number of further features will now be described.

[0046] Typically the method includes analysing the at least one characteristic and controlling the at least one operating parameter using results of the analysis. In one particular example, the processing device quantifies one or more characteristics of the output material to determine characteristic value(s) and controls the operating parameter(s) using the characteristic value(s). Quantifying characteristics in this manner allows straightforward control mechanisms to be implemented, for example, by comparing the quantified characteristic to a threshold value or range of values, adjusting the operating parameter depending on results of the comparison. This could similarly be used to allow a look-up table to be used to look up ideal operating parameters for given characteristic values.

[0047] The nature of the characteristics can vary depending on the preferred implementation. For example, when the material is processed by extruding the material, the processing device could analyse the image data to identify an extrusion edge and then determine a characteristic value by quantifying an edge straightness. Alternatively, this could involve analysing images to quantify a density of surface defects, or to identify material colours, which can in turn be used to quantify a colour homogeneity.

[0048] In one example, the images are analysed by executing a computational model, such as neural network, which has been trained to identify and / or quantify relevant material characteristics. This could be achieved, for example, by training the model using reference images of processed material having different characteristics. In one particular example, this could be achieved using a feature extractor such as a convolutional neural network or similar. As such approaches are known, these will not be described in any further detail.

[0049] The particular control strategy used can vary depending on the preferred implementation, including on whether a defined control strategy is known in advance. For example, a control strategy can be determined dynamically by determining one or more characteristics, adjusting one or more operating parameters, and then determining the effect of the change in operating parameter. This can be used to assess if the change has been effective,and in particular improved the characteristics of the processed material, which in turn can be used to inform further changes to operating parameter(s).

[0050] For example, if a change in operating parameter worsens the characteristics, different changes can be implemented, whereas if the characteristics are improved, the changes can be retained or reinforced. Thus, if increasing the processing temperature increases surface defects, then the temperature can be reduced. Conversely, if lowering the temperature reduces surface defects, further decreases in temperature could be implemented until there is no further improvement, or until other characteristics are adversely impacted.

[0051] It will be appreciated that some operating parameter changes may enhance some characteristics, while reducing others, so that a complex interplay between characteristics and parameters may need to be managed. To assist with this process, in one example, the processing device can analyse characteristic(s) by executing at least one computational model that has been trained by applying machine learning to reference characteristics derived from processed material processed using different operating parameters. The use of a machine learning trained model in this manner can help balance the different operating parameters that can be employed, thereby optimizing control of the processing equipment. The nature of such computational model(s) can vary depending on the preferred implementation, but could include the use of one or more decision trees, a random forest including multiple decision trees, a Bayes classifier, such as a Gaussian naive Bayes classifier, a multi-layer perceptron neural network, or the like. It will also be appreciated that the techniques could be applied to other machine learning approaches and subsequent resulting computational models, and the above list is therefore intended to be illustrative rather than limiting.

[0052] It will be appreciated that the characteristics that can be monitored will vary depending on the preferred implementation and the nature of the material and processing performed. Example characteristics that can be monitored when recycling plastics, such as PVC, include but are not limited to a material shape, a material structure, a material colour, material surface features, such as surface bubbles, or the like. However, this is not intended to be limiting and other characteristics could be examined.

[0053] Similarly, operating parameters of the processing equipment will also vary depending on the nature of the processing performed, but could include parameters such as a processing temperature, a processing pressure, a processing rate, amounts of processing additives used, or the like. In this regard, examples of additives that might be used include plasticizers, fire retardants, pigments, antioxidants, fillers, impact modifiers, lubricants, biocides, casting resins, colorants, processing aids, waxes, or the like.

[0054] In one particular example, the processing equipment includes a material extruder configured to extrude material in accordance with the at least one operating parameter. Such a material extruder will typically include a hopper configured to receive granular material, a screw configured to urge material through a barrel to an outlet and a heater configured to heat at least one of the barrel and the outlet. In arrangements of this form, the method can include transporting the processed material past an image sensor, for example using a conveyor belt or similar.

[0055] An example of this such an arrangement for processing granulated PVC will now be described with reference to Figure 3.

[0056] In this example, the apparatus 300 includes processing equipment in the form of an extruder 310 including a body 311 and a hopper 312, for receiving granulated PVC. The body includes a barrel 313 extending to an outlet die 314. A screw threaded auger (not shown) is provided internally within the barrel 313, and is driven by a motor 315. Heaters 313.1, 314.1 are provided allowing temperatures of the barrel 313 and outlet die 314 to be independently controlled. In use, granulated PVC received in the hopper 312 is pushed along the barrel by the auger, so that pressure and heat can be applied to the granulated material, with the resulting processed material being extruded through the outlet die 314. It will be appreciated that additional equipment can be provided allowing additives to be combined with the granulated PVC, for example adding the additives to the hopper.

[0057] The extruded material is typically provided to a transport mechanism 340, such as a conveyor belt 341 entrained around rollers 342, and supported by legs 343, allowing the extruded material to be conveyed past an image sensor 320.

[0058] In this example, the image sensor is provided in a housing 351 which straddles the conveyor belts, and which includes openings 352 allow processed material to pass therethrough. The image sensor 320 is provided in the housing 351 with a light source 321, such as a ring light being provided so that the processed material is subject to consistent illumination during imaging. This helps improve reliability of characteristic detection and in particular avoids erroneous evaluation of characteristics caused by changes in ambient illumination.

[0059] In this example, the apparatus further includes a processing system 330 including at least one microprocessor 331, a memory 332, an optional input / output device 333, such as a keyboard and / or display, and an external interface 334, interconnected via a bus 335 as shown. In this example the external interface 334 can be utilised for connecting the processing system 330 to the image sensor 320 and the processing equipment 310. For example, this can be used to provide direct connectivity to the heaters 313.1, 314.1 and drive 315, allowing processing temperatures, pressures and rates to be controlled, although this is not essential and alternatively the processing system could be connected to a separate controller. Although a single external interface 334 is shown, this is for the purpose of example only, and in practice multiple interfaces using various methods (e.g. Ethernet, serial, USB, wireless or the like) may be provided.

[0060] In use, the microprocessor 331 executes instructions in the form of applications software stored in the memory 332 to allow the required processes to be performed, for example to execute one or more computational models to analyse images, and use material characteristics to control the processing equipment. The applications software may include one or more software modules, and may be executed in a suitable execution environment, such as an operating system environment, or the like.

[0061] Accordingly, it will be appreciated that the processing system 330 may be formed from any suitable processing system, such as a suitably programmed client device, PC, or the like. In one particular example, the processing system 330 is a standard processing system such as an Intel Architecture based processing system, which executes software applications stored on non-volatile (e.g., hard disk) storage, although this is not essential. However, it will also be understood that the processing system could be any electronic processing device such as amicroprocessor, microchip processor, logic gate configuration, firmware optionally associated with implementing logic such as an FPGA (Field Programmable Gate Array), or any other electronic device, system or arrangement.

[0062] An example control process for the equipment of Figure 3 will now be described with reference to Figure 4.

[0063] In this example, at step 600 processing of material is commenced, with this typically being performed using default operating parameters. At step 610, the image sensor 320 acquires images of the processing material extruded onto the conveyor belt 341. At step 620, the processing system 330 analyses the images, for example using a feature extractor or other similar processing, in order to determine material characteristics at step 630, typically including at least an edge straightness, a colour homogeneity, and surface defect density.

[0064] In one example, image analysis is performed by first performing foreground extraction. This step separates the extruded PVC from the background in the image using Otsu's method [Nobuyuki Otsu (1979). "A threshold selection method from gray-level histograms". IEEE Trans. Sys. Man. Cyber. 9 (1): 62-66.] based on color differences. The pixels corresponding to the foreground (i.e., the extruded PVC) are extracted for processing in subsequent steps.

[0065] Following this, edge extraction and analysis can be performed. This extracts the two edges of the extruded PVC from the image, using information obtained by performing the foreground extraction. The edges are then analyzed to provide a quantitative measure of their smoothness.

[0066] Since the edges of a good product should appear as straight lines in the image, the smoothness of an edge is quantified by the deviation of the edge from the straight line. In the program, a fifth-degree polynomial instead of a straight line is fitted to the x-y coordinates of the pixels along each edge to account for lens distortion.

[0067] Following this surface defect, and in particular bubble detection is performed. This step counts the number of bubbles on the surface of the product, which can be achieved using a convolution neural network (CNN) trained via deep learning, or a conventional computer vision approach. This latter approach involves detecting edges within the foreground areausing the Canny edge detector [Canny, J., A Computational Approach To Edge Detection, IEEE Transactions on Pattern Analysis and Machine Intelligence, 8(6):679-698, 1986]. A morphological dilation operation is then applied to the result, which serves to smooth the detection result and connect disconnected edges that belong to the same bubble. Finally, the bubbles are detected by finding the contours in the smoothed edge detection result.

[0068] As a final processing step, color analysis is performed, in which the color value of each pixel is compared against the average color value of a 3x3 patch centered at the pixel. Color inconsistency is declared if the difference is larger than a pre-defined threshold.

[0069] Other processing approaches could include using a machine learning algorithm to detect the presence of large bumps / bulges on the surface of PVC products, which helps in deciding whether the temperature of the extruder is too low. The image processing software can also detect the minimum width of the PVC produce, which can be used to analyse the PVC product’s quality.

[0070] At step 640, the processing system 330 determines if the material processing is acceptable, for example by comparing quantified characteristic values to ideal value ranges. If it is determined that the characteristic values are unacceptable, for example if these fall outside the idealised ranges, one or more of the operating parameters are adjusted at step 450, with further material being processed using the updated operating parameters. This process is then repeated, allowing further monitoring to be performed until the processed material is deemed acceptable, at which point processing can continue with the current parameters, until there is a change in properties of the material being processed, at which time a change in characteristics will be detected, allowing refinement of the operating parameters to be repeated.

[0071] Thus, it will be appreciated that this approach allows feedback from images of processed material to be used to iteratively adjust operating parameters so that the material is being processed in an acceptable manner regardless of the composition of the feedstock material.

[0072] The above described arrangements can be used for processing a wide range of materials, and examples include a polymeric material, Polyvinyl Chloride (PVC), a material undergoing recycling, a granulated material, a granulated PVC material or the like.

[0073] Thus, it will be appreciated that in one specific example, the above described approach can provide a device capable of analysing PVC recyclate, with the aim adapting processing to account for differences in material composition, such as differences in existing additives, modifying the process to ensure the resulting PVC material meets a desired grade. In one example, the specific apparatus used includes an extruder, conveyer belt, camera and camera housing, analysis software and device control software creating realtime autonomous adjustments to the extruder settings.

[0074] This approach is therefore capable of analysing PVC recyclate, with the aim of identifying existing additives, and modifying the PVC recyclate to a desired grade. In a first step recyclate is extruded onto a conveyer belt where the resulting extruded recyclate is filmed. In the second step the film is analysed using image processing software to identifying the presence of existing additives, and identify changes to processing conditions to produce a desired grade of recyclate. In the third step the identified changes in conditions are fed back to the extruder and the conditions adjusted repeating steps one to three until the desired grade is achieved.

[0075] This can therefore address the difficulty in assessing the material composition of PVC to be recycled and relates to an improved methodology to autonomously grade and modify PVC recyclate to aid in processing PVC recyclate.

[0076] In one example, this is achieved using image analysis software that is designed to analyse video of extruded recycled PVC. The program was developed to detect bubbles, color inconsistencies, and rough edges of the product and the information is used to determine whether the PVC has sufficient heat stabilizer and couple this information to the extruder settings.

[0077] A graphical user interface (GUI) can be provided to allow results of the analysis to be displayed in real time, allowing the user to choose the camera to use, configure the background algorithms, and configure the saving of captured videos and analysis results. The GUI also provides the option of loading a saved video for analysis, which makes it easy to test and debug the system.

[0078] Throughout this specification and claims which follow, unless the context requires otherwise, the word “comprise”, and variations such as “comprises” or “comprising”, will be understood to imply the inclusion of a stated integer or group of integers or steps but not the exclusion of any other integer or group of integers. As used herein and unless otherwise stated, the term "approximately" means ±20%.

[0079] Persons skilled in the art will appreciate that numerous variations and modifications will become apparent. All such variations and modifications which become apparent to persons skilled in the art, should be considered to fall within the spirit and scope that the invention broadly appearing before described.

Claims

AMENDED CLAIMS received by the International Bureau on 05 November 2025 (05.11 .2025)A method of controlling equipment used in processing a material, particularly a mix of recycled polyvinyl chloride (PVC) and additives, by analysing the mix to determine its characteristics and, particularly, a deficiency in additives required for particular applications; the method involving, firstly, making one or more flat test pieces (processed material), operating parameters of the test piece forming device being progressively applied with increasing intensity over the length of a test piece and extending fully over a range of conditions in which degradation of the material is known to occur; and secondly, during forming of a test piece, the capturing of optical images of an exposed surface of it, the surface being suitably illuminated and the images captured by a suitable image sensor, the point of onset of and the degree and nature of degradation being recorded; and, thirdly, conversion into digital form of the optical images, together with known details of the polymer mix and the operating parameters of the test piece forming device throughout the forming process (digital data), and analysis of the digital data in one or more processing devices having reference to known additive deficiency-related degradation effects; analysis of the digital data allowing determination of one or more characteristics of the processed material and, thereby, control of one or more operating parameters of the equipment.The method of Claim 1 in which controlling of equipment includes control of operating parameters of a commercial polymer extrader, including a mix blending system in which additives are combined with the feedstock supplied to the polymer extruder, the mix blending system forming an integral part of the polymer extruder or being physically separate.The method of Claim 1 in which the known additive deficiency-related degradation effects are contained in an addressable reference in the form of a look-up table, knowledge base or the like.The method of Claim 1 in which test pieces are formed by extrusion.The method of Claim 1 in which test pieces are formed by rollers.The method of Claim 1 in which test pieces are formed by pressing between plates.The method of Claim 1 in which evidence of degradation includes surface perturbations, intact and ruptured bubbles, edge perturbations and colour changes.The method of Claim 1 in which the one or more processing devices analyse the digital data by executing at least one computational model, the at least one computational model being trained by applying machine learning to reference images of the processed material having different characteristics, or to reference characteristics derived from processed material processed using different operating parametersThe method of Claim 8 in which the at least one computational model includes computational models, such as neural networks or similar, which have been trained to identify and / or quantify features of interest, or a feature extractor such as a convolutional neural network or similar, or including one or more decision trees, or a Bayes classifier, or a Gaussian naive Bayes classifier or a multi-layer perceptron neural network or the like.The method of Claim 9 in which the computational model is constructed to exhibit artificially intelligent behaviour through, algorithms, simulations and models to solve complex problems, including the use of techniques such as neural networks, genetic algorithms, fuzzy logic and expert systems.The method of Claim 1 in which operating parameters of the equipment include a processing temperature, a processing pressure, a processing rate and an amount of at least one processing additive, including plasticizers, heat stabilizers, anti-oxidizers, fillers, impact modifiers, lubricants, processing aids, or the like.The method of Claim 1 in which test pieces are deposited onto and carried past the image sensor on a conveyor belt.The method of Claim 1 in which the image sensor and its illumination source source arecontained in a housing, the illumination source being configured to expose the processed material to consistent illumination.The method of Claim 1 in which the polyvinyl chloride (PVC) material is a material undergoing recycling and is granulated and the micronized to increase its surface area and obtain a uniform size.The method of Claim 1 in which the one or more processing devices form part of a controller used to control the processing equipment or are separate from but in communication with the controller and are optionally connected to the equipment directly or electronically to control any of its operating parameters.The method of Claim 1 in which the image sensor includes a camera, a CMOS sensor and associated optics or the like.The method of Claim 1 in which the one or more processing devices are microprocessorbased processing systems running applications software created for the purpose, in order to facilitate discrimination between the effects of the very large numbers of combinations of polyvinyl chloride (PVC) and additives, the applications software including an ontology in which the relationships between PVC compounds, additives to PVC compounds and the effects of PVC processing methods have been modelled.The method of Claim 1 in which images are optically captured from the lower threshold of the macro range.The method of Claim 1 in which image analysis is performed by first performing foreground extraction, this step separating a test piece from the background in the image using Otsu's method [Nobuyuki Otsu (1979). "A threshold selection method from graylevel histograms". IEEE Trans. Sys. Man. Cyber. 9 (1): 62-66] based on color differences, the pixels corresponding to the foreground being extracted for processing in subsequent steps, allowing edge extraction and analysis to be performed.The method of Claim 19 in which the two edges of a test piece are extracted from an image, using information obtained by performing the foreground extraction, the edges then being analyzed to provide a quantitative measure of their smoothness.The method of Claim 20 in which the smoothness of an edge is quantified by the deviation of the edge from a straight line, in the software a fifth-degree polynomial instead of a straight line being fitted to the x-y coordinates of the pixels along each edge to account for lens distortion.The method of Claim 7 in which bubble detection is performed by counting the number of bubbles on the surface of a test piece, achieved using a convolution neural network (CNN) trained via deep learning, or a conventional computer vision approach, the latter involving detecting edges within the foreground area using the Canny edge detector [Canny, J., A Computational Approach To Edge Detection, IEEE Transactions on Pattern Analysis and Machine Intelligence, 8(6): 679-698, 1986], a morphological dilation operation then being applied to the result, serving to smooth the detection result and connect disconnected edges belonging to the same bubble, bubbles being detected by finding the contours in the smoothed edge detection result.The method of Claim 7 in which color analysis is performed by comparing the color value of each pixel in comparison with the average color value of a 3x3 patch centered at the pixel, colour inconsistency being detected if the difference is larger than a pre-defined threshold.The method of Claim 7 in which the presence of large bumps or bulges on the surface of a test piece are detected by using a machine learning algorithm.The method of Claim 7 in which the processing system determines if the nature of the processed material is acceptable, for example by comparing quantified characteristic values to ideal value ranges, a determination that the characteristic values are unacceptable causing one or more of the operating parameters to be adjusted, with further material being processed using the updated operating parameters, the process then being repeated until the processed material is deemed acceptable, at which point commercial processing continues using the current operating parameters, until there is a change in properties of the material being processed.The method of Claim 25 in which feedback from images of processed material are used to iteratively adjust operating parameters so that the processed material is acceptable.The method of Claim 1 in which a system is created capable of analysing polyvinyl chloride (PVC) recyclate and adapting processing to account for differences in material composition, such as differences in existing additives, modifying the process to ensure the resulting PVC material meets a desired grade.The method of Claim 1 in which polyvinyl chloride (PVC) recyclate is analysed with the objective of identifying existing additives and modifying the PVC recyclate to a desired grade and to autonomously grade and modify PVC recyclate to aid in the processing of PVC recyclate.The method of Claim 1 in which the analysis is employed to determine whether the polyvinyl chloride (PVC) has sufficient heat stabilizer additive.The method of Claims 1, 8, 9 and 10 in which a graphical user interface (GUI) is provided to allow results of the analysis to be displayed in real time, allowing a user to choose the type of image sensor, to configure the background algorithms, to configure the saving of captured images and analysis results, and of loading saved images for analysis to facilitate testing and debugging of the system.

Images