Method and device for monitoring heap of material

Abstract

To monitor a heap of a material in a recycling station to reduce the number of false alarms.SOLUTION: A device comprises a circuit (120) configured to perform a region of interest function (132), a reference spatial characteristic setting function (134), a region of interest adjustment function (136), and a temperature monitoring function (138). The region of interest function is configured to define a region of interest covering a heap of a material in a thermal video stream of a scene captured by a thermal video camera. The reference spatial characteristic setting function is configured to determine a spatial characteristic of the heap of the material from pixels within a region of interest of a video frame of the thermal video stream, and to set the determined spatial characteristic as a reference spatial characteristic. The temperature monitoring function is configured to monitor temperature measure in the video stream within the region of interest over time, and generate an alarm event when the temperature measure exceeds a predetermined threshold.SELECTED DRAWING: Figure 1

Description

[Technical Field] 【0001】 The present invention relates to monitoring heaps of material using thermal video cameras and devices configured to perform such monitoring. [Background technology] 【0002】 Thermal cameras are excellent devices for monitoring various scenes, especially those in which the temperature of one or more objects in the scene should be monitored. One such example is monitoring a heap of material at a recycling station. By monitoring the temperature of the heap of material, an alarm regarding an increase in temperature can be triggered. However, there is generally a problem of false alarms due to machines with internal combustion engines moving around at the recycling station. Therefore, there is a need to reduce the number of false alarms. Summary of the Invention 【0003】 The object is to mitigate, alleviate or eliminate one or more of the above-identified deficiencies and disadvantages in the art singly or in any combination, or to solve at least the problems mentioned above. 【0004】 According to a first aspect, there is provided a method for monitoring a heap of material, the method including: defining a region of interest covering the heap of material in a thermal video stream of a scene captured by a thermal camera, determining a reference spatial characteristic of the heap of material from pixel data within the region of interest from video frames of the thermal video stream, adjusting the region of interest such that the region of interest encompasses an area in the video stream that exhibits the sample spatial characteristic above a spatial characteristic threshold, monitoring temperature measurements in the video stream over time within the region of interest, and generating an alarm event when the temperature measurement exceeds a predetermined threshold. 【0005】 This method for monitoring a heap of material allows the region of interest to be dynamically adjusted to cover the heap of material, even as the heap changes size over time, is moved, etc. This allows for the avoidance of false alarms, since only the heap may be covered by the region of interest, even as the heap changes size or is moved over time. This is achieved by adjusting the region of interest so that it encompasses areas in the video stream that exhibit sample spatial characteristics above a spatial characteristic threshold. The spatial characteristic threshold should be high enough so that non-heap areas are excluded from the heap. The spatial characteristic threshold may be, for example, 60% of the reference spatial characteristic. The reference spatial characteristic of the heap should be viewed as an optical signature for the heap. The spatial characteristic may be, for example, a measure of the spatial frequency of the wavelengths of the thermal image data comprising the video stream. The spatial frequency may be entropy, edge density, or another spatial-frequency-related metric. Spatial frequencies (especially at thermal wavelengths) typically exhibit a wide distribution for heaps of material, while areas covering buildings or ground generally exhibit uniform spatial frequencies. Thus, by adjusting the region of interest based on the spatial characteristics of areas in the scene shown by the thermal video stream, the method allows an operator to track moved or altered heaps of material without having to update the region of interest over time. 【0006】 The method may further include adjusting the region of interest such that regions in the video stream exhibiting sample spatial properties below a spatial property threshold are excluded by the adjusted region of interest. 【0007】 The temperature measure may be one or more of an absolute value of the temperature and a rate of change of the absolute value of the temperature. 【0008】 The method may further include updating the reference spatial characteristics. The updating of the reference spatial characteristics may be triggered by one or more of the ambient temperature of the thermal camera, the contrast of the video stream, weather data associated with the scene, the season, and the time of day. Updating the reference spatial characteristics may account for changes in the heap of material over time. For example, the size of the material in the heap may fluctuate over time due to the addition or removal of material to the heap. 【0009】 The reference and sample spatial properties can be calculated for groups of pixels containing 4x4 to 8x8 pixels, which significantly reduces the amount of processing power required in the process. 【0010】 The method may further include defining multiple regions of interest in the video stream, each region of interest covering a specific heap of material; determining a specific reference spatial characteristic for each specific heap of material from pixel data within each specific region of interest covering the specific heap of material from video frames of the video stream; adjusting the specific regions of interest so that each specific region of interest encompasses an area in the video stream exhibiting a sample spatial characteristic within 90% to 110% of the respective specific reference spatial characteristic; setting a specific predetermined threshold for a temperature measure for each specific region of interest; monitoring the specific temperature measure in the video stream within each specific region of interest over time; and generating an alarm event when the specific temperature measure exceeds the specific predetermined threshold. Thus, the spatial frequency distribution within the thermal wavelength range for a heap containing a specific material can be used as an optical signature for each specific heap of material. This allows different types of temperature triggers to be set for different regions of interest, i.e., different heaps of material. 【0011】 According to a second aspect, there is provided a non-transitory computer-readable storage medium having stored thereon instructions for performing a method according to the first aspect when executed on a device having processing capability. 【0012】 According to a second aspect, a device for monitoring a heap of material is provided. The device comprises circuitry configured to perform a region of interest function, a reference spatial property setting function, a region of interest adjustment function, and a temperature monitoring function. The region of interest function is configured to define a region of interest covering the heap of material in a thermal video stream of a scene captured by a thermal video camera. The reference spatial property setting function is configured to determine spatial properties of the heap of material from pixels within the region of interest in a video frame of the thermal video stream and set the determined spatial properties as reference spatial properties. The region of interest adjustment function is configured to determine respective sample spatial properties for regions in the thermal video stream and adjust the region of interest so that the region of interest includes regions in the thermal video stream that exhibit sample spatial properties above a spatial property threshold. The temperature monitoring function is configured to monitor temperature measurements in the video stream over time within the region of interest and generate an alarm event when the temperature measurements exceed a predetermined threshold. As described above, the spatial property threshold may be, for example, 60% of the reference spatial property. 【0013】 The region of interest adjustment function may be further configured to adjust the region of interest such that regions in the video stream exhibiting sample spatial properties below a spatial property threshold are excluded by the adjusted region of interest. 【0014】 The temperature measure may be one or more of an absolute value of the temperature and a rate of change of the absolute value of the temperature. 【0015】 The reference spatial property setting function may be further configured to update the reference spatial property by determining a spatial property of the heap of material from pixels within a current region of interest of a video frame of the video stream when triggered by one or more of an ambient temperature of the thermal camera, a contrast of the video stream, weather data associated with the scene, a season, and a time of day, and setting the determined spatial property as an updated reference spatial property. 【0016】 The reference spatial property setting function and the region of interest adjustment function may be configured to determine the reference spatial property and the sample spatial property as one or more of image entropy and density of edges, respectively. 【0017】 The set reference spatial property function and adjust region of interest function may be configured to calculate the reference spatial property and the sample spatial property, respectively, for groups of pixels including 4x4 to 8x8 pixels. 【0018】 The region of interest function may be configured to define multiple regions of interest in the video stream, each region of interest covering a specific heap of material. The reference spatial property setting function may be configured to determine a specific reference spatial property for each specific heap of material from pixels within each respective specific region of interest of a video frame of the thermal video stream. The region of interest adjustment function may be configured to adjust the specific regions of interest such that each specific region of interest encompasses an area in the video stream exhibiting a sample spatial property within 90% to 110% of the respective specific reference spatial property. The region of interest adjustment function may further be configured to set a specific predetermined threshold for a temperature measure for each region of interest. The temperature monitoring function may be configured to monitor a specific temperature measure in the video stream over time within each specific region of interest and generate an alarm event when the specific temperature measure exceeds a specific predetermined threshold for the specific region of interest. 【0019】 The device may further comprise an image sensor configured to capture thermal video data of a thermal video stream of the scene. 【0020】 The above-mentioned features of the method also apply to the second and third aspects, where applicable, and in order to avoid undue repetition, reference is made to the above. 【0021】 Further scope of applicability of the present invention will become apparent from the detailed description given hereinafter. However, it should be understood that the detailed description and specific examples, while indicating preferred embodiments of the invention, are given by way of example only, since various changes and modifications within the scope of the invention will become apparent to those skilled in the art from the detailed description. 【0022】 Thus, it should be understood that the systems and methods described may vary, and that the present invention is not limited to the specific components of such systems or acts of such methods. It should also be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not limiting. It should be noted that, as used in this specification and the appended claims, the articles "a," "an," "the," and "said" are intended to mean that there are one or more of an element, unless the context clearly dictates otherwise. Thus, for example, reference to "a device" or "the device" may include several devices, etc. Furthermore, the words "comprising," "including," "including," and similar phrases do not exclude other elements or steps. 【0023】 These and other aspects of the present invention will now be described in more detail with reference to the accompanying drawings, which should not be considered limiting, but rather are used for purposes of explanation and understanding. 【0024】 As shown in the figures, the sizes of layers and regions may be exaggerated for purposes of explanation and, therefore, are given to show the general structure, and like reference numbers refer to like elements throughout. [Brief explanation of the drawings] 【0025】 [Figure 1] FIG. 1 shows a device for monitoring a heap of material. [Figure 2] FIG. 1 is a block diagram of a method for monitoring a heap of material. [Figure 3A] 1 shows a heap of material shown by a thermal video camera at different time instances. [Figure 3B] 1 shows a heap of material shown by a thermal video camera at different time instances. [Figure 3C] 1 shows a heap of material shown by a thermal video camera at different time instances. DETAILED DESCRIPTION OF THE INVENTION 【0026】 The present invention now will be described more fully hereinafter with reference to the accompanying drawings, in which presently preferred embodiments of the invention are shown. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided for thoroughness and completeness, and to fully convey the scope of the invention to those skilled in the art. 【0027】 FIG. 1 illustrates a device 100 for monitoring a heap of material. The device 100 is typically implemented in a thermal video camera. However, the device may be configured to receive a thermal video stream from the thermal video camera. In the following, the device 100 is described as being implemented in a thermal video camera. The device 100 in the form of a thermal video camera includes a thermal image sensor 112. The thermal image sensor 112 is configured to capture thermal video data. The thermal video data is typically pixel-based data. The device 100 further includes a video image pipeline 114. The video image pipeline 114 is configured to process the thermal video data into video frames of a thermal video stream. The thermal video camera covers a field of view 111 that depicts a scene. In this implementation, the thermal video camera is configured to depict a scene including one or more heaps 190 of material. The material in each of the one or more heaps 190 of material may be of different types, such as wood waste, waste metal, plastic waste, paper waste, electronic waste, etc. Typically, the thermal video camera is set up to monitor a scene covering at least a section of the recovery center. The thermal video camera is also typically set up to monitor the temperature of the heap of material 190 and to generate an alarm when a temperature increase above a threshold is monitored. The threshold may be a measure of absolute temperature in the heap 190. Alternatively, or in combination, the threshold may be a rate of change of temperature in the heap 190. The threshold may be set differently for different heaps of material; i.e., a heap consisting primarily of wood may have a lower threshold (in absolute temperature) than a heap consisting primarily of glass. 【0028】 Device 100 further comprises circuitry 120. Circuitry 120 is configured to provide overall control of the functions and operations of device 100. Circuitry 120 may include a processor 121, such as a central processing unit (CPU), microcontroller, or microprocessor. Processor 121 is configured to execute program code stored in memory 130 to perform the functions and operations of device 100. 【0029】 Memory 130 may be one or more of a buffer, flash memory, a hard drive, removable media, volatile memory, non-volatile memory, random access memory (RAM), or another suitable device. In a typical configuration, memory 130 may include non-volatile memory for long-term data storage and volatile memory that serves as system memory for circuit 120. Memory 130 may exchange data with circuit 120 via a data bus. There may also be associated control lines and address buses between memory 130 and circuit 120. 【0030】 The functions and operations of device 100 may be embodied in the form of executable logic routines (e.g., lines of code, software programs, etc.) stored on a non-transitory computer-readable medium (e.g., memory 130) of device 100 and executed by circuitry 120 (e.g., using processor 121). Furthermore, the functions and operations of device 100 may be standalone software applications or may form part of a software application that performs additional tasks related to device 100. The functions and operations described may be thought of as methods that the corresponding parts of the device are configured to perform. Also, while the functions and operations described may be implemented in software, such functions may equally be performed via dedicated hardware or firmware, or some combination of hardware, firmware, and / or software. 【0031】 The circuit 120 is configured to execute a region of interest function 132. The region of interest function 132 is configured to define a region of interest covering the heap of material 190 in a thermal video stream of a scene captured by the thermal video camera. The region of interest function 132 may be configured to receive user input to define the region of interest. For example, a user may mark an area that constitutes the region of interest on a display showing the scene covered by the thermal video camera. Alternatively, or in combination, a machine learning model may be used to define the region of interest. The region of interest defined by the region of interest function 132 does not necessarily need to cover the entire heap of material 190. It is sufficient that a portion of the heap 190 is covered by the region of interest. Preferably, the region of interest covers an area of ​​at least 10×10 pixels in the thermal video data of a frame of the thermal video stream. In combination with defining a region of interest for the heap, the region of interest function 132 may be further configured to attribute a temperature threshold for the heap 190. The temperature threshold is the threshold at which the device 100 is configured to issue an alarm, see temperature monitoring function 138 described below. 【0032】 The circuit 120 is further configured to execute a reference spatial characteristic setting function 134. The reference spatial characteristic setting function 134 is configured to determine spatial characteristics of the heap of material from pixels within a region of interest of one or more video frames of the thermal video stream. The pixels within the region of interest used to determine the spatial characteristics of the heap are preferably pixels that are sufficiently inside the heap. The reference spatial characteristic setting function 134 is further configured to set the determined spatial characteristics as reference spatial characteristics. The reference spatial characteristics of the heap are to be viewed as an optical signature for the heap 190. The spatial characteristics may be, for example, a measure of spatial frequency in thermal wavelengths of the thermal image data constituting the thermal video stream. The spatial frequency may be entropy, edge density, or another spatial frequency-related metric. By way of non-limiting example, the spatial frequency may be determined by applying a Laplacian filter and / or a Sobel filter to pixel data of the thermal image data. The pixel data to which the Laplacian filter and / or Sobel filter is applied may be grouped into groups of neighboring pixels. Such a group of neighboring pixels may contain 4x4 to 8x8 pixels. Spatial frequencies (especially at thermal wavelengths) generally exhibit a wide distribution for a heap of material, while areas covering buildings or ground generally exhibit uniform spatial frequencies. The spatial frequency distribution within the thermal wavelength range for a heap containing a particular material can be used as an optical signature for the heap of material. 【0033】 The circuit 120 is further configured to execute a region of interest adjustment function 136. The region of interest adjustment function 136 is configured to determine respective sample spatial characteristics for a region in the thermal video stream. The region of interest adjustment function 136 is generally configured to operate on pixel data of thermal image data of an image frame subsequent to the image frame used to establish the reference spatial characteristics. The region of interest adjustment function 136 may be executed periodically. The frequency of the periodically executed region of interest adjustment function 136 may be fixed. For example, the region of interest adjustment function 136 may be executed periodically within a period on the order of one minute to one hour. The frequency of the periodically executed region of interest adjustment function 136 may be random. Alternatively, the region of interest adjustment function 136 may be executed in response to movement detected in the region of interest. Movement in the region of interest is a strong indicator that the heap is being processed, i.e., either material has been added or removed, either of which provides a reason for adjustment of the region of interest. Movement can be easily detected using thermal camera analytics. Each sample spatial characteristic for a region in the thermal video stream is preferably determined in the same manner as the reference spatial characteristic. Thus, the sample spatial frequency may be determined by applying a Laplacian filter and / or a Sobel filter to the pixel data of the thermal image data. The pixel data to which the Laplacian filter and / or Sobel filter is applied may be grouped into groups of neighboring pixels. Such groups of neighboring pixels may include 4x4 to 8x8 pixels. The region of interest adjustment function 136 may be configured to set a lower threshold for the area to configure the heap of material to be an area that covers at least a 3x3 coherent group of neighboring pixels. The region of interest adjustment function 136 is configured to adjust the region of interest so that regions in the thermal video stream exhibiting sample spatial characteristics above the spatial characteristic threshold are included by the region of interest.Expressed differently, the region of interest adjustment function 136 is configured to adjust the region of interest such that regions in the thermal video stream exhibiting sample spatial characteristics below a spatial characteristic threshold are excluded by the adjusted region of interest. Thus, by adjusting the region of interest based on spatial characteristics of regions in the scene shown by the thermal video stream, this setup allows an operator to track moved or changed heaps of material without having to update the region of interest over time. The spatial characteristic threshold may be set to be 60% of the reference spatial characteristic. 【0034】 The circuit 120 is further configured to perform a temperature monitoring function 138. The temperature monitoring function 138 is configured to monitor a temperature measurement in the thermal video stream over time and within a region of interest. The temperature measurement may be one or more of an absolute temperature value and a rate of change for the absolute temperature value. The temperature monitoring function 138 is further configured to generate an alarm event when the temperature measurement exceeds a predetermined threshold. 【0035】 The reference spatial property setting function 134 may be further configured to update the reference spatial property by determining a spatial property of the heap of material from pixels within a current region of interest of a video frame of the video stream and setting the determined spatial property as an updated reference spatial property. The updating of the reference spatial property may be configured to occur periodically, for example, once a day or once a week. Alternatively, the updating of the reference spatial property may be triggered by one or more of the ambient temperature of the thermal video camera, the contrast of the thermal video stream, weather data associated with the scene, the season, and the time of day. The device 100 may include a temperature sensor 140 for determining the ambient temperature of the thermal video camera. The device 100 may be connected to a weather service that provides weather data associated with the scene. The device 100 may include a transceiver 150, and the device is configured to communicate with other devices or services, such as a weather service, via the transceiver 150. 【0036】 Therefore, when there is a change in the surrounding environment, the reference spatial characteristics often need to be updated. For example, if the ambient temperature is changing, the reference spatial characteristics should be updated. The ambient temperature can be estimated, for example, by determining the sensor temperature. Alternatively, the ambient temperature can be estimated by determining temperature measurements from pixels outside the region of interest. Alternatively, external weather information can be used. 【0037】 Additionally, the contrast of a thermal video stream may be affected by, for example, weather, time of day, etc. A change in contrast above a set threshold may trigger an update of the reference spatial characteristics. Furthermore, an update of the reference spatial characteristics may be triggered when the contrast within the region of interest compared to the region outside the region of interest begins to diverge. According to one example, the region of interest is set during bad weather, such as rain or snow. As the weather improves, the difference between the contrast within the region of interest compared to the region outside the region of interest begins to differ more. In this case, it is time to update the reference spatial characteristics. A condition for performing an update may be that the difference between the contrast within the region of interest compared to the region outside the region of interest increases by 10% or more. Another condition that may be set is that the absolute difference between the contrast within the region of interest compared to the region outside the region of interest exceeds a threshold, for example, 50%. Furthermore, when initially setting up a region of interest, if the absolute difference between the contrast within the requested region of interest compared to the region outside the region of interest falls below a threshold (e.g., below a 25% difference in absolute numbers), a warning is issued informing the user that the initial region of interest cannot be set. 【0038】 Furthermore, when the contrast within the region of interest and outside the region of interest begins to become more equal, i.e., when the difference between these contrasts falls below a threshold (e.g., below a 25% difference in absolute numbers), no update of the reference spatial characteristics occurs. 【0039】 Also, when a thermal camera is calibrated, the total scene temperature dynamics can be estimated, and if this temperature dynamics changes by more than a set threshold number of degrees, this can be a trigger to update the reference spatial characteristics. 【0040】 2, a method 200 for monitoring a heap of material will be described. Some or all of the steps of method 200 may be performed by device 100 described above. However, it will be equally understood that some or all of the steps of method 200 may be performed by one or more other devices having similar functionality. Method 200 includes the following steps. The steps may be performed in any suitable order. 【0041】 A region of interest covering the heap of material is defined in a thermal video stream of the scene captured by the thermal camera (S202). This step is further illustrated in FIG. 3A, which shows a frame of a thermal video stream of a scene including a heap of material. The region of interest 192 covering the heap of material 190 defined by step S202 is shown in this figure. Step S202 of defining a region of interest covering the heap of material has been described in more detail above with respect to region of interest function 132. To avoid undue repetition, reference is made to this description. 【0042】

[0023] From one or more video frames of the thermal video stream, determine (S204) a reference spatial property of the heap of material from pixel data within the region of interest. The step S204 of determining the reference spatial property is described in more detail above with respect to the reference spatial property setting function 134. To avoid excessive repetition, reference is made to this description. 【0043】 The region of interest is adjusted (S206) so that regions in the video stream exhibiting sample spatial characteristics above the spatial characteristic threshold are included by the region of interest. This step may alternatively or in combination be expressed as adjusting the region of interest so that regions in the video stream exhibiting sample spatial characteristics below the spatial characteristic threshold are excluded by the adjusted region of interest. The step of adjusting the region of interest S206 was described in more detail above with respect to the region of interest adjustment function 136. The spatial characteristic threshold may be 60% of the reference spatial characteristic. Reference is made to this description to avoid excessive repetition. From the above description of the region of interest adjustment function 136, it will be understood that the adjusting step S206 is performed on one or more frames of the thermal video stream following the frame from which the reference spatial characteristic was determined. Two different examples of such subsequent frames are shown in FIGS. 3B and 3C. In FIG. 3B, the heap of material 190 is moving. Therefore, the region of interest 192 has been adjusted in position to move with the heap of material. Additionally, in FIG. 3B, the heap of material 190 has been reduced in size. 3C, the heap of material 190 is divided into two heaps 190a, 190b. Accordingly, the region of interest is divided into two regions of interest 192a, 192b. In the example described above, each region of interest is shown as a rectangular region of interest, but it will be appreciated that the region of interest may take any shape that reflects the shape of the heap of material. 【0044】 A temperature measurement in the video stream is monitored over time within the region of interest (S208). The temperature measurement may be one or more of an absolute value of the temperature and a rate of change of the absolute value of the temperature. 【0045】 If the temperature measurement exceeds a predetermined threshold, an alarm event is generated (S210). 【0046】 The method may further include updating the reference spatial characteristics (S207). The updating of the reference spatial characteristics may be triggered by one or more of the ambient temperature of the thermal camera, the contrast of the video stream, weather data associated with the scene, the season, and the time of day. 【0047】 Those skilled in the art will appreciate that the present invention is by no means limited to the preferred embodiments described above. On the contrary, many modifications and variations are possible within the scope of the appended claims. 【0048】 For example, the method may be performed on multiple different specific heaps of material. Each specific heap of material is generally a heap of material that is different from other heaps, e.g., a heap of metal material, a heap of paper material, a heap of plastic material, etc. If so, the method may include: defining multiple regions of interest in the video stream, each region of interest covering a specific heap of material; determining, from video frames of the video stream, specific reference spatial characteristics for each specific heap of material from pixel data within each respective specific region of interest that covers the specific heap of material; adjusting the specific regions of interest so that each specific region of interest encompasses an area in the video stream that exhibits a sample spatial characteristic within 90% to 110% of the respective specific reference spatial characteristic; setting a specific predetermined threshold for a temperature measure for each specific region of interest; monitoring a specific temperature measure in the video stream within each specific region of interest over time; and generating an alarm event if the specific temperature measure exceeds a specific predetermined threshold. 【0049】 Additionally, variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims. [Explanation of symbols] 【0050】 100 devices 112 Thermal Image Sensor 114 Video Image Pipeline 120 circuits 121 processors 130 memory 132 Region of Interest Function 134 Reference space characteristic setting function 136 Region of interest adjustment function 138 Temperature monitoring function 140 Temperature Sensor 150 Transceiver 190 Heap 192 Areas of Interest

Claims

[Claim 1] A method used in a device having processing capacity for monitoring a material heap (190), wherein the method is In the video stream of the scene captured by the thermal camera, define the region of interest encompassing the heap of material, Determining a reference spatial characteristic of the heap of material from the pixel data within the region of interest, from the video frames of the video stream, wherein the reference spatial characteristic is a measure of the spatial frequency of the thermal image data. Setting a spatial characteristic threshold for the aforementioned reference spatial characteristics, Adjusting the region of interest such that the region of interest includes the region in the video stream that exhibits sample spatial characteristics exceeding the spatial characteristic threshold, Monitoring the temperature measurement in the video stream over time and within the region of interest, When the temperature measurement exceeds a predetermined threshold, an alarm event is generated. A method that includes this. [Claim 2] The method according to claim 1, further comprising adjusting the region of interest such that any region in the video stream exhibiting sample spatial characteristics below the spatial characteristic threshold is excluded from the adjusted region of interest. [Claim 3] The method according to claim 1, wherein the temperature measure is one or more of the absolute value of the temperature and the rate of change of the absolute value of the temperature. [Claim 4] The method according to claim 1, further comprising updating the reference spatial characteristics, wherein the update of the reference spatial characteristics is triggered by one or more of the ambient temperature of the thermal camera, the contrast of the video stream, weather data related to the scene, season, and time of day. [Claim 5] The method according to claim 1, wherein the reference spatial characteristics and the sample spatial characteristics are one or more of image entropy and edge density. [Claim 6] The method according to claim 1, wherein the reference spatial characteristics and the sample spatial characteristics are calculated for groups of pixels including 4x4 to 8x8 pixels. [Claim 7] Defining multiple regions of interest in the aforementioned video stream, wherein each region of interest encompasses a specific heap of material, From the video frames of the video stream, determine a specific reference spatial characteristic for each particular heap of the material from the pixel data within each specific region of interest encompassing the specific heap of the material, Adjusting the specific regions of interest such that each of the specified regions of interest includes a region in the video stream that exhibits sample spatial characteristics ranging from 90% to 110% of the respective specified reference spatial characteristics, For each specific region of interest, a specific predetermined threshold value is set for the temperature measurement. Monitoring specific temperature measurements in the video stream over time and within each specific region of interest, When the aforementioned specific temperature measurement exceeds the aforementioned specific predetermined threshold, an alarm event is generated. The method according to claim 1, further comprising: [Claim 8] A non-temporary computer-readable storage medium storing instructions for carrying out the method according to any one of claims 1 to 7 when executed on a device having processing capabilities. [Claim 9] A device for monitoring a material heap (190), wherein the device is A region of interest function (132) configured to define a region of interest encompassing the heap (190) of material in a thermal video stream of a scene captured by a thermal video camera, Determining the spatial properties of the heap of a material from pixels within the region of interest of the video frame of the thermal video stream, wherein the spatial properties are measures of the spatial frequency of the thermal image data. The determined spatial characteristics are set as the reference spatial characteristics, Setting a spatial characteristic threshold for the aforementioned reference spatial characteristics, A reference spatial characteristic setting function (134) configured to perform the following, Determining the sample-space characteristics of each region in the thermal video stream, Adjusting the region of interest such that the region of interest includes a region in the thermal video stream that exhibits sample spatial characteristics exceeding the spatial characteristic threshold. A region of interest adjustment function (136) configured to perform the following: Monitoring the temperature measurement in the thermal video stream over time and within the region of interest, When the temperature measurement exceeds a predetermined threshold, an alarm event is generated. A temperature monitoring function (138) configured to perform the following: Circuit (120) configured to perform the following: A device equipped with the following features. [Claim 10] The device according to claim 9, wherein the temperature measure is one or more of the absolute value of the temperature and the rate of change of the absolute value of the temperature. [Claim 11] The device according to claim 9 or 10, wherein the reference spatial characteristic setting function is further configured to update the reference spatial characteristic by triggering one or more of the ambient temperature of the thermal video camera, the contrast of the thermal video stream, weather data related to the scene, season, and time of day, determining the spatial characteristics of the heap of material from pixels in the currently region of interest of the video frames of the thermal video stream, and setting the determined spatial characteristics as the updated reference spatial characteristic. [Claim 12] The device according to claim 9 or 10, wherein the reference spatial characteristic setting function and the region of interest adjustment function are configured to determine the reference spatial characteristic and the sample spatial characteristic as one or more of image entropy and edge density, respectively. [Claim 13] The device according to claim 9 or 10, wherein the reference spatial characteristic setting function and the region of interest adjustment function are configured to calculate the reference spatial characteristic and the sample spatial characteristic, respectively, for groups of pixels including 4x4 to 8x8 pixels. [Claim 14] The region of interest function (132) is configured to define a plurality of regions of interest in the thermal video stream, where each region of interest encompasses a specific heap of material. The reference spatial characteristic setting function (134) is configured to determine a specific reference spatial characteristic for each specific heap of material from pixels within each specific region of interest of each video frame of the thermal video stream, The region of interest adjustment function (136) is configured to adjust each specific region of interest such that each specific region of interest includes a region in the thermal video stream that exhibits sample spatial characteristics of 90% to 110% of each specific reference spatial characteristic, and to set a specific predetermined threshold for the temperature measurement for each specific region of interest. The aforementioned temperature monitoring function (138) is, Monitoring specific temperature measurements in the thermal video stream over time and within each specific region of interest, When the aforementioned specific temperature measurement exceeds the aforementioned specific predetermined threshold for the aforementioned specific region of interest, an alarm event is generated. The device according to claim 9 or 10, configured to perform the following: [Claim 15] The device according to claim 9 or 10, further comprising an image sensor configured to capture thermal video data of the thermal video stream of the scene.

Images