System and procedure for monitoring conveyor belts

Abstract

Detection system (10) for conveyor belt temperature abnormalities, comprising: an infrared line scanner (12) comprising one or more infrared detectors and arranged to scan the surface of a conveyor belt (16) in a direction perpendicular to the direction of movement of the conveyor belt (16) and to generate thermal imaging data depending on infrared radiation detected by the infrared detector; a processing unit (14) configured to process the thermal imaging data in order to identify temperature abnormalities in the material moving on the surface of the conveyor belt (16), and a two-dimensional arrangement of infrared detectors arranged to receive infrared radiation from the surface of the conveyor belt (16) and to generate thermal imaging data depending on the received infrared radiation; and wherein the processing unit (14) is further configured to process the thermal imaging data in order to identify temperature abnormalities in the material moving on the surface of the conveyor belt (16).

Description

AREA OF INVENTION

[0001] This invention relates to conveyor belt monitoring and in particular a system and a method for detecting temperature abnormalities in materials moving over conveyor belts. BACKGROUND OF THE INVENTION

[0002] Many industries require temperature monitoring of materials moving along conveyor belts. Traditionally, this monitoring is performed using a single spot or dot pyrometer pointed at the belt. Such techniques are of limited value, as they provide only a rough indication of the temperature in the area on which the spot is focused. If the spot size is too large, small temperature abnormalities, especially those smaller than the spot size, may be missed. Alarms associated with dot pyrometers are limited to a warning upon momentary local exposure to an upper / lower limit, frequently resulting in false or missed alarms.

[0003] US 2015 / 0095254A1 concerns a baggage drop-off and check-in system for airline flights. The system comprises a conveyor equipped with a static or dynamic scale comprising multiple spaced-apart load cells, a calculating device designed to compare the scale output with the permitted baggage weights, and a substantially horizontal frame element positioned above the first conveyor belt at a distance from the top of the first conveyor belt corresponding to the maximum permitted baggage height, thereby forming a physical barrier against oversized baggage. The system allows for the verification of the presence of living beings on the conveyor belt and compliance with the baggage handling system requirements.The system can further include a baggage drop-off device to enable automatic baggage check-in by a passenger, including a device for paying for excess baggage and / or excess weight. The invention further comprises a method for checking in baggage using the baggage drop-off system according to the invention.

[0004] US 2003 / 0197126A1 relates to a device and a method for detecting impurities in a material in which infrared rays or a variety of specific wavelength components of infrared rays are applied to a material on a conveyor belt, the respective reflection intensities of the specific wavelength components reflected by the material are measured, the measured reflection intensities and the reflection intensities of specific wavelength components inherent in the material are compared, and impurities in the material are detected according to the result of the comparison.

[0005] US 2013 / 0278771A1 concerns systems and methods that use small infrared imaging modules to monitor various components of a vehicle wheel assembly. For example, a vehicle-mounted system may include one or more infrared imaging modules, a processor, memory, a display, a communication module, and a vehicle speed sensor. The vehicle-mounted system may be attached to, installed in, or otherwise integrated into a vehicle with one or more wheel assemblies. The one or more infrared imaging modules may be configured to acquire thermal images of desired parts of the wheel assemblies. Various thermal imaging analyses and profiling can be performed on the acquired thermal images to determine the operational status of different wheel assembly components and to detect anomalies.Based on the detected condition and anomalies, monitoring information can be generated and displayed to the driver or other occupants of the vehicle in real time.

[0006] US 2007 / 0108288A1 relates to a method and apparatus, as well as similar methods and apparatus, which provide a method for extracting barcode information from surfaces where the codes are formed by either depressions or protrusions. It includes the extraction of DataMatrix 2D barcode patterns and the subsequent analysis of the content of markings on forged steel parts that exhibit surface defects which render current state-of-the-art readers ineffective. It enables the analysis of images derived from the surface morphology itself, rather than merely the contrast in a standard camera image produced by typical directional or specifically non-directional lighting.

[0007] DE 695 07 832 T2 relates to a sorting device that has a conveyor belt or equivalent mechanism for moving particles at a speed sufficient to generate a particle stream in the air, whereby the particles can be sorted in such a way that impermissible material can be removed. Sorting is carried out by a primary scanning system that analyzes the light reflected by the particles in the stream across several wavelength ranges. Ejectors for removing particles from the stream are arranged downstream of the scanning system and are triggered in response to signals received by the scanning system. An additional scanning system is also included to detect the presence of material in the stream. If a gap is detected in a specific area, the analysis of that area by the primary scanning system and the corresponding activation of the ejectors are prevented.

[0008] DE 195 19 861 A1 relates to a method and a device for detecting foreign bodies in a stream of particulate material, in particular tobacco, comprising an optical scanner in combination with a camera system with beam splitter and filters, and a computer for processing the camera signals to determine whether foreign bodies are present. If foreign bodies are detected, a deflection device redirects the stream of contaminated material onto a conveyor belt.

[0009] The present invention is therefore based on the objective of providing a method and a recognition system, each of which is suitable to enrich the prior art.

[0010] The problem is solved by the features of the independent claim and the dependent claims. The subclaims each contain further developments of the disclosure. SUMMARY OF THE INVENTION

[0011] According to a first aspect of the invention, a detection system for conveyor belt temperature abnormalities is provided with the features of claim 1.

[0012] According to a second aspect of the invention, a conveyor belt temperature abnormality detection system is provided, comprising a two-dimensional arrangement of infrared detectors arranged to receive infrared radiation from the surface of a conveyor belt and to generate thermal imaging data depending on the received infrared radiation; and a processing unit configured to process the thermal imaging data in order to identify temperature abnormalities in the material moving on the surface of the conveyor belt.

[0013] According to a third aspect of the invention, a method for detecting temperature abnormalities in material transported on a conveyor belt is provided, comprising the features of claim 17.

[0014] According to a fourth aspect of the invention, a method for detecting temperature abnormalities in material transported on a conveyor belt is provided, the method comprising: scanning the surface of the conveyor belt with an infrared line scanner in a direction perpendicular to the direction of movement of the conveyor belt; generating thermal imaging data based on infrared radiation received by the infrared line scanner from the material moving on the conveyor belt; processing the thermal imaging data to identify temperature abnormalities in material moving on the surface of the conveyor belt.

[0015] According to a fifth aspect of the invention, a computer-readable storage medium is provided which has computer-readable instructions stored thereon which, when executed by a processor in conjunction with a thermal line scanner, perform the above-mentioned process steps. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention will now be described by reference to a non-limiting example and the accompanying drawings, wherein: Fig. 1 a schematic diagram of a conveyor belt abnormality detection system according to an embodiment of the present invention; Fig. 2 a perspective view of a line scanner of the in Fig. Figure 1 shows the conveyor belt abnormality detection system, which displays the line scanner's viewing profile relative to the conveyor belt; Fig. 3 is a flowchart that describes a process for identifying temperature abnormalities in the system of Fig. 1 recorded thermal images; Fig. 4. A flowchart is a process for filtering one of the system's components. Fig. 1 captured image represents; Fig. 5 a graphical representation of the system of Fig. 1 captured thermal images before and after filtering using the process of Fig. 4 is; Fig. 6. A flowchart is a process for identifying abnormalities in the system of Fig. 1 captured images; and Fig. 7 a graphical representation of one of the system of Fig. 1 of the captured thermal image, in which abnormalities were marked by dotted lines. DESCRIPTION OF EXECUTIONS OF THE INVENTION

[0017] Embodiments of the present invention overcome the disadvantages associated with conventional spot pyrometer measurement techniques by using an infrared line scanner to repeatedly scan the entire width of a conveyor belt as material moves along its surface. Data received from successive line scans can be combined to form a thermal image of the belt and the material on it. Temperature abnormalities present in the thermal image can then be identified and presented to a user, allowing for appropriate action to be taken.

[0018] Embodiments of the present invention are described below primarily with reference to the monitoring of cement clinker moving on a conveyor belt, typically to and from process kilns. In cement processing plants, it is desirable to monitor localized hot spots present in the clinker. However, it will be understood that in other applications it may be desirable to measure abnormally low-temperature areas, i.e., cold spots, in material moving on a conveyor belt. Embodiments of the present invention can therefore also be used to identify areas of below-average temperature (or "cold spots") in material moving on a conveyor belt.

[0019] Fig. 1 and Fig. Figure 2 schematically depicts a conveyor belt abnormality detection system 10 according to an embodiment of the present invention. The conveyor belt abnormality detection system 10 comprises an infrared scanner 12 and a processing unit 14 associated with and connected to the scanner 12. A display 26 and an input device 24, such as a keyboard or a touchscreen, are coupled to the processing unit 14.

[0020] In use, the scanner 12 can be positioned above a conveyor belt 16 to scan the belt 16 across the width of the belt 16 in a “y” direction (in Fig. 2) to scan the scanning line 18, which is oriented perpendicular to the "x" longitudinal axis of the tape. In the Fig. 1 and Fig. In the embodiments shown in 2, the conveyor belt 16 operates to transport material along the length of the belt in the “x” direction, so that material moves from left to right across the side.

[0021] The infrared line scanner 12, including one or more infrared detectors, is configured to repeatedly scan the tape 16 across its width, such that the focal spot of the infrared detector(s) sweeps across the surface of the tape 16 and any material on it. The tape 16 can be scanned by the line scanner 12 either by physically moving the detector or by shifting its focus across the tape using a device such as a rotating mirror, thereby achieving a linear sequence of measurements across the tape 16. A two-dimensional thermal profile of the tape 16 (including any material on it) can then be generated by combining the data received by the infrared detector(s) over several line scans. The system 10 can include a single infrared detector or multiple detectors.In the case of a single detector, the detector can be moved scanning to receive infrared radiation at the focus on the material, as discussed above. In the case of multiple detectors, a linear array of detectors can be arranged to receive radiation emitted over the entire width of band 16. The multiple detectors can be moved scanning according to the procedures described above for a single detector, or alternatively, they can be stationary but scan electronically.

[0022] In any case, the infrared line scanner 12 can have a variable scanning angle, preferably between 15° and 120°, which allows the scanner 12 to be mounted at a suitable height above the tape 16 while still capturing images of the entire tape width. Furthermore, the scanner 12 is ideally oriented such that the angle of incidence of the focus detector on the tape is 90° relative to the 'y'-plane of the tape 16, i.e., the surface of the tape along its length. It will be understood that other angles of incidence relative to the tape 16 are within the scope of the invention. However, the angle should preferably be no less than 60° relative to the 'y'-plane of the tape 16. Beyond this point, the effective emissivity of the scanned material may decrease, particularly for reflective materials moving across the tape 16, such as metals.If the material moving over tape 16 has an uneven surface, this will further lead to shadows in the processed image due to an oblique viewing angle of less than 60°.

[0023] Data received by the infrared line scanner 12 is transmitted to the processing unit 14 via one or more buses. The processing unit 14 can be implemented as part of the line scanner 12 or separately from it. The processing unit 14 can be implemented in software running on a PC, for example, or alternatively in hardware using one or more digital signal processors (DSPs) or application-specific integrated circuits (ASICs). Software running on the processing unit 14 can include instructions which, when executed, cause the line scanner 12 and the infrared detector to operate as described below. The processing unit 14 can include one or more local or remote storage devices for storing infrared data received from the line scanner 12.Instructions executed by processing unit 14 can also be stored in the local or remote storage facilities associated with processing unit 14.

[0024] In addition to receiving infrared data, the processing unit 14 can also receive speed data from a belt speed sensor 22 coupled to the conveyor belt 16. Using this information, the processing unit 14 can generate an image of the belt 16 that is insensitive to distortions caused by speed fluctuations of the belt 16 (the faster the belt 16 moves, the further apart the scan lines are assembled, and vice versa).

[0025] A user can enter one or more criteria for the detection of abnormalities via the input device 24, as described in more detail below. Furthermore, the thermal images of the band 16, once generated, can be displayed on the display 26 connected to the processing unit 14.

[0026] It will be evident that the image generated by the processing unit 14 is a continuous image whose length increases with each scan of the tape 16. Accordingly, for the purpose of detecting abnormalities on the tape, the processing unit 14 can store a limited number of line scans in memory, thereby reducing the system's memory requirements. Alternatively, or in addition, all line scan data received by the processing unit can be stored for later analysis in a manner known in the prior art.

[0027] The process for capturing and analyzing line scan data received from the line scanner 12 by the processing unit 14 is now described with reference to Fig. As described in section 3, in step 30, the processing unit 14 receives scan lines from the thermal line scanner. This data can be processed in real time or stored in one or more buffer memories and / or permanent storage for later analysis. Furthermore, in step 32, the processing unit 14 can also receive real-time tape speed data from the tape speed monitor 22. In step 34, using the tape speed data to calculate the scan line spacing required to eliminate image distortion, the processing unit 14 generates a thermal image from the scan lines, with each scan line being spaced by a distance proportional to the speed of the tape 16 at the time that specific scan was performed.

[0028] Once the adjusted thermal image has been assembled, the image can be filtered in step 36 to remove high-frequency spatial information. Then, in step 38, temperature abnormalities in the image can be identified that relate to hot and / or cold areas present in the material moving across band 16.

[0029] Referring to Fig. 4 will now be the one in Fig. The filter step shown in Figure 3 is described. First, in step 40, a filter window is generated. The dimensions of the filter window can be set by a user of the conveyor belt abnormality detection system 10, for example, using the input device 24. In some embodiments, the filter window dimensions are set to be equal to the minimum size of an abnormality that a user wishes the system 10 to detect. For example, a user can input a length and width of the minimum size of an abnormality to be identified into the processing unit 14, and this data is then used to determine the length and width of the generated filter window. Abnormalities with dimensions smaller than those of the filter window are thus removed from the filtered image.

[0030] Once generated, the filter window can be moved across the width of the composite thermal image. At each position in the scan, the average of the pixels in the filter window is calculated (step 44). The center pixel in the filter window at each position in the scan is then set to the average of the pixels in that filter window, calculated for that position. Once the filter reaches the end of the image width, in step 48 the filter window is moved to the next unfiltered section of the image, and the process returns to step 42, moving the filter window across the width of the composite thermal image. The process is repeated until no further image data needs to be filtered, for example, when the conveyor belt stops or the system is switched off.

[0031] The result is a filtered thermal image in which high-frequency spatial information has been removed. In other words, small anomalies that are hotter than the upper temperature limit or colder than the lower temperature limit, but smaller than the filter window, have been removed. These small anomalies are therefore not visible when the filtered image is scanned for temperature abnormalities. Larger anomalies, however, remain visible. Furthermore, in the case of a temperature abnormality that is at least as large as the minimum size being scanned for, the presence of some pixels that lie outside the temperature limit does not prevent the detection of this abnormality, as long as the remaining pixels forming the abnormality are sufficiently far above the limit to compensate and push the average value for the filter window above the limit.A simple temperature limit can be applied to the filtered thermal image without generating false alarms due to abnormalities that are smaller than the set user limit.

[0032] Fig. Figure 5 shows thermal images before (left) and after (right) the preceding filtering process, in which edges of abnormalities are blurred and small temperature abnormalities have been removed.

[0033] It will be apparent that the preceding filtering step is not strictly necessary. However, performing this step increases the overall efficiency of System 10, as abnormalities smaller than the user-set threshold, such as areas with very few high- or low-value pixels, will not be detected in the filtered image and thus will not be available for subsequent abnormality searches. Any pixel in the filtered image that exceeds the threshold should be part of a genuine abnormality.

[0034] It will be apparent that alternative methods for low-pass filtering are known in this field and can be used instead of the aforementioned method. For example, a weighted average can be calculated within the filter window, so that pixels located in the center contribute more to the output value than pixels at the edges or corners of the window. In this case, weight values ​​per pixel could be calculated, with the result being similar to the preceding averaging method. The selection of a weighted or unweighted filter may depend on factors such as whether the abnormalities being sought are expected to have a uniform temperature, and whether defects in the material on the tape are to be detected or whether the tape itself is to be protected from temperature-related damage. Such methods are not outside the scope of the present invention.

[0035] If we now refer back to Fig. 3, after the thermal image has been assembled and optionally filtered in step 36, the thermal image can be processed in step 38 to identify pixels and groups of pixels that correspond to areas of material moving over the conveyor belt 16 that have an abnormally high (or low) temperature. Fig. Section 6 describes a process flow for identifying such abnormalities. In step 50, pixels in the thermal image are identified that have a value exceeding a predetermined temperature threshold. This temperature threshold can be set by a user and can be an absolute temperature. Alternatively, the threshold can be set as the difference between a maximum pixel value and an average pixel value for the thermal image. In some embodiments, the rate of temperature change in an identified area can be monitored, and thresholds can be set based on these characteristics. Additionally, or alternatively, the temperature difference between different parts of a linear target can be used. For example, the temperature difference between the centerline and the edges of strip steel in a rolling mill can be used as a threshold for temperature abnormalities.

[0036] Pixels in the image identified as having a value exceeding the threshold are then grouped in step 52. This grouping can be achieved by considering neighboring pixels as part of the same abnormality. To consolidate the number of detected features, pixels that are close together but not touching can also be considered as belonging to the same abnormality and thus grouped as well. This can be achieved by reducing the resolution of the filtered image through peak picking or valley picking of either the hottest or the coldest pixels in a small area, thereby creating a single pixel with reduced resolution.Immediately adjacent pixels in this reduced-resolution image can then belong to a single consolidated abnormality, which can be combined with other reduced-resolution pixels and / or pixels that are immediately adjacent to any of the higher-resolution pixels forming the reduced-resolution pixels.

[0037] In step 54, for each suspected abnormality detected—that is, each group of pixels suspected of forming the same abnormality—dimensions are calculated and compared to a threshold value, also set by a user. If the suspected abnormality meets the threshold criteria, in step 56 the suspected abnormality is considered to satisfy the threshold criteria, and data relating to this abnormality can be saved for further analysis. Conversely, if the suspected abnormality is considered smaller than required by the threshold criteria, the abnormality is disregarded in step 58. This process is repeated for each pixel or group of pixels with a value exceeding the threshold criterion.If the low-pass filtering step was performed before this step, it is not necessary to check whether the abnormality has sufficient dimensions, since abnormalities smaller than the threshold are removed during the filtering step. In such circumstances, steps 54, 56, and 58 can be skipped.

[0038] During the detection and analysis of the thermal imaging data by the processing unit 14, some or all of the generated images can be displayed on the display 26 coupled to the processing unit 14. For example, the display 26 can show images before and after low-pass filtering. Additionally, or alternatively, after the detection of an abnormality in the thermal image, one or more markers can be superimposed on the image to indicate to the user where abnormalities occur on band 16 and which of them meet the limit criteria, etc. Fig. Figure 7 shows a sample thermal image generated by processing unit 14, which can be displayed to the user. Temperature changes are indicated by color changes in the image (in the illustration in Figure 7). Fig. 7 shown in black and white); wherein darker areas of the image represent cold areas of the conveyor belt and lighter areas of the image represent hot areas of the conveyor belt. In this embodiment, the square boxes were drawn around abnormalities that proved to meet the requirements of the user-set limit criteria. These limits are set to identify hot spots 62 in the material moving over the conveyor belt 16. However, as explained above, embodiments of the present invention can also or alternatively identify cold spots in thermal images.

[0039] In addition to visually identifying temperature abnormalities in a thermal image, the system can include one or more alarms configured to trigger when a temperature abnormality is detected that meets user-defined threshold criteria. These alarms can be audible or visual, such as a siren or flashing light, to alert a user to an abnormality. Alarms can be presented to a user via the display 26, for example, as on-screen messages. Such messages can also be visually linked to an abnormality shown in the thermal image, even on the screen.

[0040] In response to an alarm condition, the processing unit 14 can generate an output signal to trigger an external event, such as stopping the belt's movement. The output signal can be in the form of a 4-20 mA output signal or an OPC (Object Linking and Embedding for Process Control) compatible signal. OPC can also be used by the processing unit 14 to couple with additional sensors and monitors (e.g., the conveyor belt speed monitor 22) to receive additional data from the conveyor belt 16 and any other associated devices. OPC can also be used to couple the conveyor belt monitoring system 10 with other process control and instrumentation systems used in a plant where the monitoring system 10 is installed. Furthermore, the monitoring system 10 can be coupled with a knowledge management system, such as ABB's Knowledge Manager.

[0041] Different alarms can be implemented for different threshold values ​​or criteria. Alternatively, or in addition, an alarm criterion can be set that requires the presence of a predetermined number of detected abnormalities in an area of ​​interest, e.g., the last sampled section of the tape up to a user-specified distance from the sampling point.

[0042] An alarm database can be maintained to store details of each alarm event, including recorded conveyor data and any other data received by the processing unit at the time of the alarm. This allows alarm events to be reviewed and analyzed later.

[0043] In the embodiments described above, a single set of limit criteria is applied to the received image data, meaning that only temperature abnormalities that meet this single set of limit criteria are identified. However, in other embodiments, multiple filters with different criteria can be applied to the same input data, allowing temperature abnormalities with different characteristics to be identified in the same image. For example, a first filter can be set to identify hot spots / areas in material moving across conveyor belt 16, and a second filter can be configured to identify cold spots in material moving across conveyor belt 16.

[0044] In addition to applying multiple filters with different criteria, some designs allow the area of ​​the thermal image to be preset. Fig. Figure 7 shows a pair of boundary lines 64 used to mark the area of ​​the image in which abnormalities are to be identified. Areas outside the boundaries 64 are not analyzed. Such boundaries can be set, for example, if the scan width of the line scanner 12 is wider than the belt. The boundary lines can prevent areas outside the belt 16, such as metal side rails, electric motors, etc., from being included in the abnormality identification process. Such features can have undesirable effects on calculations performed during image processing. By cropping out unwanted areas of the image, these features do not affect the result of the boundary value analysis performed by the processing unit 14. Furthermore, in some embodiments, two conveyor belts can be imaged in parallel using a single line scanner.Scanner 12 would then scan the width of the two tapes. In this case, it may be desirable to analyze the material moving on each tape individually. Limits can be set around the area of ​​the tape requiring analysis, and the [data] can be [data] in [data]. Fig. 3, Fig. 4 and Fig. The processes described in section 6 can only be carried out in this area for certain predefined criteria.

[0045] In some system configurations, the line of sight to the line scanner may be obstructed. In all of the embodiments described above, one or more additional line scanners can be positioned to ensure a completely unobstructed view of the conveyor belt 16. For example, multiple scanners can be configured to view the belt 16 from different angles. Infrared data generated by each scanner can then be processed and combined to represent a single stream of scan data. Scan data from the optimally positioned line scanner (i.e., the line scanner with the best view of the belt 16) can then be selected and used for thermography to avoid obstructions and achieve the highest resolution.Furthermore, in the case of blockages that cause disturbances in the view, spot pyrometers can be used in addition to or as an alternative to one or more additional line scanners.

[0046] In the embodiments described above, a line scanner with one or more infrared detectors is provided to scan the width of the tape. In each of the embodiments described above, the line scanner can be replaced by a thermal imaging camera (such as an infrared camera) configured to thermographically image material moving along the tape. The thermal imaging camera can include a two-dimensional array of infrared detectors directed at the tape. Accordingly, a two-dimensional scan or image of the tape can be performed, and the received image data can be used by the processing unit 14 to generate the thermal image of the material on the tape.The thermal imaging camera can transmit data in the form of individual images to the processing unit 14 for processing, or the processing unit 14 can itself generate individual images and then process them.

[0047] In some of the embodiments described above, a belt speed monitor 22 provides the processing unit 14 with a belt speed reading. Although providing a belt speed monitor 22 is preferable, in other embodiments, instead of using the belt speed monitor 22, an analysis of the path of hot or cold spots in the thermal image could be used to measure the belt speed. For example, the distance traveled by an anomaly in a given period, and thus the belt speed at any given time, can be calculated in real time, provided that at least one anomaly is present in the image. The anomaly used to determine the belt speed need not exceed the threshold required for identification or for triggering an alarm event. Reference symbol list 10 Recognition system 12 infrared scanners 14 processing units 16 Conveyor belt 18 scanning lines 22 tape speed meters 24 Input device 26 Display 30-58 steps 62 places 64 boundary lines

Claims

Detection system (10) for conveyor belt temperature abnormalities, comprising: an infrared line scanner (12) having one or more infrared detectors and arranged to scan the surface of a conveyor belt (16) in a direction perpendicular to the direction of movement of the conveyor belt (16) and to generate thermal imaging data depending on infrared radiation detected by the infrared detector; a processing unit (14) configured to process the thermal imaging data to identify temperature abnormalities in the material moving on the surface of the conveyor belt (16); and a two-dimensional arrangement of infrared detectors arranged to receive infrared radiation from the surface of the conveyor belt (16) and to generate thermal imaging data depending on the received infrared radiation;and wherein the processing unit (14) is further configured to process the thermal imaging data in order to identify temperature abnormalities in the material moving on the surface of the conveyor belt (16). Detection system (10) for conveyor belt temperature abnormalities according to claim 1, wherein the processing of the thermal image data comprises: receiving a plurality of scan lines from the thermal line scanner, wherein the scan lines comprise a plurality of pixels, each having a pixel value corresponding to a temperature of the material detected at that pixel; assembling a thermal image of the conveyor belt (16) from the plurality of scan lines, wherein the thermal image is formed from the plurality of pixels; and identifying groups of pixels in the thermal image that have abnormal pixel values ​​corresponding to temperature abnormalities in the material. Detection system (10) for conveyor belt temperature abnormalities according to claim 1, wherein the processing of the thermal image comprises: receiving the thermal image data from the arrangement of infrared detectors comprising a plurality of pixels, each having a pixel value corresponding to a temperature value of the material detected by that pixel; assembling a thermal image of the conveyor belt (16) from the plurality of pixels; and identifying groups of pixels in the thermal image that have abnormal pixel values ​​corresponding to temperature abnormalities in the material. Detection system (10) for conveyor belt temperature abnormalities according to claim 2 or 3, wherein prior to identifying the groups of pixels the processing unit (14) is configured to filter the image to remove high-frequency spatial information from the image. Detection system (10) for conveyor belt temperature abnormalities according to claim 4, wherein the filtering comprises: sweeping a filter window over the thermal image; and calculating a filtered pixel value for the pixel centered in the filter window at each sampling position of the filter window, wherein the filtered pixel value is equal to one of the following: a) the value of a mean value of the pixels in the filter window; and b) the value of a weighted average of the pixels in the filter window. Detection system (10) for conveyor belt temperature abnormalities according to claim 5, wherein the dimensions of the filter window correspond to the minimum dimensions of the abnormalities to be detected by the conveyor belt (16). Detection system (10) for conveyor belt temperature abnormalities according to claim 5 or 6, wherein the weighting of the weighted average is such that pixels located closer to the center of the filter window contribute more to the filtered pixel value than pixels located further from the center of the filter window. Detection system (10) for conveyor belt temperature abnormalities according to claims 2 to 7, wherein the identification comprises: selecting first pixels in the thermal image that have a pixel value exceeding a first predetermined limit temperature; and grouping adjacent or nearly adjacent selected first pixels. Detection system (10) for conveyor belt temperature abnormalities according to claim 8, wherein the identification comprises: selecting second pixels in the thermal image that have a pixel value exceeding a second predetermined limit temperature; and grouping adjacent or nearly adjacent selected second pixels. Detection system (10) for conveyor belt temperature abnormalities according to claim 8 or 9, wherein the identification further comprises determining whether each group of first pixels has a size greater than a first predetermined limit size and / or determining whether each group of second pixels has a size greater than a second predetermined limit size. Detection system (10) for conveyor belt temperature abnormalities according to one of claims 2 to 10, further comprising an input for receiving a specification of the conveyor belt speed from the conveyor belt (16), wherein the thermal image is composed depending on the received conveyor belt speed specification. Detection system (10) for conveyor belt temperature abnormalities according to one of the preceding claims, further comprising a display (26) configured to display the thermal image data and one or more indications of the identified temperature abnormalities in the thermal image. Detection system (10) for conveyor belt temperature abnormalities according to claim 12, wherein the one or more indications comprise lines drawn around the abnormalities in the thermal image. Detection system (10) for conveyor belt temperature abnormalities according to one of the preceding claims, further comprising an input device (24) for receiving limit value criteria for identifying the temperature abnormalities. Detection system (10) for conveyor belt temperature abnormalities according to one of the preceding claims, wherein the infrared detector is a heat detector or a photodetector and / or wherein the infrared detector operates in the waveband of near-infrared, mid-infrared and far-infrared. Detection system (10) for conveyor belt temperature abnormalities according to one of the preceding claims, when dependent on claim 1 or 2, wherein the line scanner comprises a linear arrangement of infrared detectors arranged to electronically scan the width of the conveyor belt (16). A method for detecting temperature abnormalities in material transported on a conveyor belt (16), comprising: scanning the surface of the conveyor belt (16) in a direction perpendicular to the direction of movement of the conveyor belt (16) with an infrared line scanner (12); generating thermal imaging data based on infrared radiation received by the infrared line scanner (12) from the material moving on the conveyor belt (16); processing the thermal imaging data to identify temperature abnormalities in the material moving on the surface of the conveyor belt (16); directing a two-dimensional array of infrared detectors at the surface of a conveyor belt; generating thermal imaging data based on the received infrared radiation; and processing the thermal imaging data to identify temperature abnormalities in the material moving on the surface of the conveyor belt (16). The method of claim 17, wherein the processing comprises: receiving a plurality of scan lines in the generated thermal image data, wherein the scan lines comprise a plurality of pixels having pixel values; assembling a thermal image of the conveyor belt (16) from the plurality of scan lines, wherein the thermal image is formed from the plurality of pixels; and identifying groups of pixels in the thermal image that have abnormal pixel values ​​corresponding to temperature abnormalities in the material. Detection method for conveyor belt temperature abnormalities according to claim 17, wherein processing the thermal image comprises: receiving the thermal image data from the arrangement of detectors, wherein the thermal image data comprises a plurality of pixels, each having a pixel value corresponding to a temperature value of the material detected by that pixel; assembling a thermal image of the conveyor belt (16) from the plurality of pixels; and identifying groups of pixels in the thermal image that have abnormal pixel values ​​corresponding to temperature abnormalities in the material. Method according to claim 18 or 19, further comprising filtering the thermal image to remove high-frequency spatial information. The method of claim 20, wherein the filtering comprises: sweeping a filter window over the thermal image; and calculating a filtered pixel value for the pixel centered in the filter window at each sampling position, wherein the filtered pixel value is equal to one of the following: a) the value of the mean of the pixels in the filter window; and b) the value of a weighted average of the pixels in the filter window. Method according to claim 21, wherein the dimensions of the filter window correspond to the minimum dimensions of the abnormalities to be detected by the conveyor belt (16). Method according to claim 22, wherein the weighting of the weighted average is such that pixels located closer to the center of the filter window contribute more to the filtered pixel value than pixels located further from the center of the filter window. Method according to any one of claims 17 to 23, wherein the identification comprises: selecting first pixels in the thermal image that have a pixel value exceeding a first predetermined limit temperature; and grouping adjacent or nearly adjacent selected first pixels. The method of claim 24, wherein the identification further comprises: selecting second pixels in the thermal image that have a pixel value exceeding a second predetermined limit temperature; and grouping adjacent or nearly adjacent selected second pixels. Method according to claim 24 or 25, further comprising determining whether each group of first pixels has a size greater than a first predetermined limiting size and / or determining whether each group of second pixels has a size greater than a second predetermined limiting size. Method according to one of claims 18 and 19 to 26, further comprising receiving a specification of the belt speed from the conveyor belt (16), wherein the thermal image is composed depending on the received conveyor belt speed specification. Method according to any one of claims 17 to 27, further comprising receiving limit criteria to identify temperature abnormalities. Method according to any one of claims 17 to 27, further comprising displaying the thermal imaging data together with an indication of the identified temperature abnormalities on a display (26). Method according to claim 29, wherein the one or more indications comprise lines drawn around the abnormalities in the thermal image. A computer-readable storage medium comprising computer-readable instructions stored thereon which, when executed by a processor in conjunction with a thermal line scanner or a two-dimensional arrangement of infrared detectors, perform the steps according to any one of claims 17 to 30.

Images