Description:TECHNICAL FIELD
[0001] The present invention relates to object classification and particle size estimation, and more particularly, to systems and methods for determining fragment (e.g., ore particles) distribution and optimization of at least one downstream operation of the fragments.
BACKGROUND
[0002] Mining and excavation involve extracting valuable minerals or other geological materials from the earth. Further, the ore particles are subjected to various mineral processing operations during excavation. However, estimating the particle size distribution (PSD) of the ore particles before downstream operations is crucial in the mining industry. In general, the particle size distribution (PSD) estimation of the objects (e.g., ore particles) is crucial in understanding the characteristics of the mined material, as it influences processing methods and the efficiency of subsequent steps in the production process.
[0003] The PSD of the ore particles is determined using various PSD estimation techniques. Some PSD estimation techniques are laser diffraction, sieve analysis, Electrical Sensing Zone, and ultrasonic extinction. However, the conventional PSD estimation techniques are time-consuming and possess one or more limitations such as a lack of accuracy in PSD estimation due to irregularly shaped particles, poor calibration factor, and the like. In addition, some of the conventional estimation techniques (e.g., electrical sensing zones) are limited to electrically conductive materials. In such cases, foreign objects such as shovel teeth, drill pipes, tyre, wood, and other scrap metal that are mixed with the ore particles during the dumping process are further proceeded to the downstream operations. As a result, these foreign objects can cause damage to the equipment (e.g., crushers) which may result in downtime and incur repair costs. Furthermore, in some of the PSD conventional estimation techniques, continuous and real-time monitoring of the PSD during the extraction process is often limited.
[0004] Therefore, there is a need for systems and methods for real-time monitoring of the PSD of the ore particles and suitably grading the ore particles for further downstream processing of the ore particles that can overcome one or more limitations stated above, in addition to providing other technical advantages.
SUMMARY
[0005] This summary is provided only for the purpose of introducing the concepts presented in a simplified form. This is not intended to identify essential features of the claimed invention or limit the scope of the invention in any manner.
[0006] In order to solve the foregoing problem and to provide other advantages, one aspect of the present invention is to provide a system for determining fragment distribution. The system includes a communication interface and a processing circuitry communicably coupled to the communication interface. The communication interface is configured to receive one or more image frames including raw materials. Further, the processing circuitry is configured to at least identify a plurality of fragments of the raw materials in the one or more image frames based at least on at least one artificial intelligence (AI) model. The processing circuitry is configured to determine a fragment size of each of the plurality of fragments identified in the one or more image frames based at least on implementing image processing techniques. Further, the processing circuitry is configured to compute a particle size distribution (PSD) metric of the plurality of fragments based at least on aggregating the fragment size of the each of the plurality of fragments identified in the one or more image frames.
[0007] In an aspect, the one or more image frames are captured while the raw materials are loaded in one of at least one dumper load and a raw material loading equipment.
[0008] In an aspect, the particle size distribution (PSD) metric of the plurality of fragments identified in the one or more image frames corresponds to the particle size distribution (PSD) metric of the at least dumper load.
[0009] In an aspect, the processing circuitry is further configured to at least determine at least one quality metric associated with the plurality of fragments in the at least one dumper load based at least on the particle size distribution (PSD) metric of the plurality of fragments in the at least one dumper load. The at least one quality metric is indicative of an overall blast efficiency of the raw materials obtained in an excavation process.
[0010] In an aspect, the processing circuitry is further configured to at least facilitate grading of the plurality of fragments in the at least one dumper load based at least on the particle size distribution (PSD) metric of the plurality of fragments in each of the at least one dumper load.
[0011] In an aspect, the processing circuitry is further configured to at least define a fragment region for each of the plurality of fragments of the raw materials in the one or more image frames based at least on applying one or more object segmentation techniques by the at least one artificial intelligence (AI) model. Further, the processing circuitry is configured to compute the fragment size of each of the plurality of fragments identified in the one or more image frames based at least on identifying at least one fragment dimension element associated with each of the plurality of fragments in the one or more image frames and applying a calibration factor.
[0012] In an aspect, the at least one artificial intelligence (AI) model is trained with training data. The training data includes historical image data of fragments obtained during the excavation process, boundary of each fragment in the historical image data, spatial locations of each fragment, and estimated particle size distribution (PSD) values of the fragments in the historical image data.
[0013] In an aspect, the processing circuitry is further configured to at least transmit the particle size distribution (PSD) metric of at least one dumper load to a central interface unit through the communication interface, and store data related to the fragment size of each of the plurality of fragments, the particle size distribution (PSD) metric associated with each dumper load in a database communicably coupled to the processing circuitry.
[0014] In an aspect, the system further includes an imaging module mounted to one of a raw material loading equipment and communicably coupled to the processing circuitry through the communication interface. The imaging module configured to at least receive a control signal from the raw material loading equipment configured to unload the raw materials to the at least one dumper load. The control signal is transmitted to the imaging module at the onset of unloading the raw materials to the at least one dumper load in each batch. Further, in response to receipt of the control signal, the imaging module captures the one or more image frames for a predetermined time for each batch of the raw materials being loaded to the at least one dumper load.
[0015] In another aspect, a computer-implemented method for determining fragment distribution is disclosed. The computer-implemented method performed by a processing circuitry includes receiving one or more image frames including raw materials. The method includes identifying a plurality of fragments of the raw materials in the one or more image frames based at least on at least one artificial intelligence (AI) model. Further, the method includes determining a fragment size of each of the plurality of fragments identified in the one or more image frames based at least on implementing image processing techniques. The method further includes computing a particle size distribution (PSD) metric of the plurality of fragments based at least on aggregating the fragment size of the each of the plurality of fragments identified in the one or more image frames.
[0016] In an aspect, the one or more image frames are captured while the raw materials are loaded in one of at least one dumper load and a raw material loading equipment.
[0017] In an aspect, the particle size distribution (PSD) metric of the plurality of fragments identified in the one or more image frames corresponds to the particle size distribution (PSD) metric of the at least dumper load.
[0018] In an aspect, the method includes determining at least one quality metric associated with the plurality of fragments in the at least one dumper load based at least on the particle size distribution (PSD) metric of the plurality of fragments in the at least one dumper load. The at least one quality metric is indicative of an overall blast efficiency of the raw materials obtained in an excavation process.
[0019] In an aspect, the method includes facilitating grading of the plurality of fragments in the at least one dumper load based at least on the particle size distribution (PSD) metric of the plurality of fragments in each of the at least one dumper load.
[0020] In an aspect, the method includes defining a fragment region for each of the plurality of fragments of the raw materials in the one or more image frames based at least on applying one or more object segmentation techniques by the at least one artificial intelligence (AI) model. Further the method includes computing the fragment size of each of the plurality of fragments identified in the one or more image frames based at least on identifying at least one fragment dimension element associated with each of the plurality of fragments in the one or more image frames and applying a calibration factor.
[0021] In an aspect, the method includes the at least one artificial intelligence (AI) model is trained with training data. The training data includes historical image data of fragments obtained during the excavation process, boundary of each fragment in the historical image data, spatial locations of each fragment, and estimated particle size distribution (PSD) values of the fragments in the historical image data.
[0022] In an aspect, the method includes transmitting the particle size distribution (PSD) metric of at least one dumper load to a central interface unit through the communication interface, and storing data related to the fragment size of each of the plurality of fragments, the particle size distribution (PSD) metric associated with each dumper load in a database communicably coupled to the processing circuitry.
[0023] In an aspect, the method includes the one or more image frames are captured by an imaging module communicably coupled to the processing circuitry. The imaging module captures the one or more image frames for a predetermined time for each batch in response to receipt of a control signal from a raw material loading equipment at the onset of unloading the raw materials to the at least one dumper load in each batch.
BRIEF DESCRIPTION OF THE DRAWINGS
[0024] For understanding exemplary embodiments of the present disclosure, reference is now made to the following descriptions taken in connection with the accompanying figures:
[0025] Figure 1 illustrates an environment related to at least some example embodiments of the present disclosure;
[0026] Figure 2 illustrates a simplified block diagram representation of a system to determine real-time fragment distribution for grading the raw materials, in accordance with an embodiment of the present disclosure;
[0027] Figures 3A and 3B illustrate a schematic representation of an image frame depicting fragments of a dumper load in a dumper truck, in accordance with an embodiment of the present disclosure;
[0028] Figure 4A illustrates a graphical representation depicting a particle size distribution (PSD) metric of the dumper load, in accordance with an embodiment of the present disclosure;
[0029] Figure 4B illustrates a graphical representation depicting a count distribution of the fragments in a dumper load, in accordance with an embodiment of the present disclosure; and
[0030] FIG. 5 illustrates a flow diagram of a computer-implemented method to determine real-time fragment distribution for grading the raw materials in each dumper load, in accordance with an embodiment of the present disclosure.
[0031] The figures referred to in this description depict embodiments of the disclosure for the purposes of illustration only. One skilled in the art will readily recognize from the following description that alternative embodiments of the systems and methods illustrated herein may be employed without departing from the principles of the disclosure described herein.
DESCRIPTION
[0032] In the following description, for purposes of explanation, numerous specific details are set forth in order to provide a broad understanding of the present disclosure. It will be apparent, however, to one skilled in the art that the present disclosure can be practiced without these specific details. In other instances, systems and methods are shown in block diagram form only in order to avoid obscuring the present disclosure.
[0033] Reference in this specification to “one embodiment” or “an embodiment” means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present disclosure. The appearance of the phrase “in one embodiment” in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Moreover, various features are described which may be exhibited by some embodiments and not by others. Similarly, various requirements are described which may be requirements for some embodiments but not for other embodiments.
[0034] Moreover, although the following description contains many specifics for the purposes of illustration, anyone skilled in the art will appreciate that many variations and/or alterations to said details are within the scope of the present disclosure. Similarly, although many of the features of the present disclosure are described in terms of each other, or in conjunction with each other, one skilled in the art will appreciate that many of these features can be provided independently of other features. Accordingly, this description of the present disclosure is set forth without any loss of generality to, and without imposing limitations upon, the present disclosure.
[0035] Various exemplary embodiments of the present disclosure are explained with references to Figure 1 to Figure 5.
[0036] Figure 1 illustrates an environment (100) related to at least some example embodiments of the present disclosure. Although the environment (100) is presented in one arrangement, other arrangements are also possible where the parts of the environment (100) (or other parts) are arranged or interconnected differently. The environment (100) is depicted to include at least one dumper truck such as a dumper truck (102a) and a dumper truck (102b). In general, the dumper truck is a heavy-duty vehicle that is designed for the transportation of bulk materials such as coal, iron ore, etc. Further, the environment (100) includes a raw material loading equipment (104). The raw material loading equipment (104) may operate to unload the raw materials to the dumper trucks (102a; 102b) until the predefined payload capacity of the dumper trucks (102a; 102b).
[0037] In the illustrated embodiment, the dumper trucks (102a; 102b) are used to transport raw materials (e.g., ore particles) and unload the raw materials. It is to be noted that the raw materials are obtained as a result of a blasting operation during the excavation process in the mining sectors. The raw materials correspond to an unprocessed state of a mineral resource as it is extracted from the earth. The raw materials in the mining sectors are generally referred to as Run of Mine (ROM). Further, the raw material loading equipment (104) is configured to unload the excavated raw materials to the dumper trucks (102a; 102b) based on their payload capacity. The raw materials loaded into each of the dumper trucks (102a; 102b) correspond to at least one dumper load (collectively referring to a dumper load (116a) and a dumper load (116b)). Some non-limiting examples of the raw material loading equipment (104) may include, but are not limited to, shovels, excavators, and belt loaders.
[0038] Further, the environment (100) includes an imaging module (106), a system (110), a central interface unit (114), and a database (112). The imaging module (106) may be configured to capture real-time image frames of the raw materials to analyze the size and shape of the particles. Some non-limiting examples of the imaging module (106) may include High-speed cameras, digital cameras, infrared cameras, 3-dimensional (3D) imaging cameras, hyperspectral cameras, Light detection and ranging (LIDAR) cameras, and the like. The imaging module (106) may be communicably coupled to the system (110) through a communication interface (or a network (108)).
[0039] Various entities in the environment (100) may connect to a network (108) in accordance with various wired and wireless communication protocols, such as Transmission Control Protocol and Internet Protocol (TCP/IP), User Datagram Protocol (UDP), 2nd Generation (2G), 3rd Generation (3G), 4th Generation (4G), 5th Generation (5G) communication protocols, Long Term Evolution (LTE) communication protocols, or any combination thereof. In some instances, the network (108) may include a secure protocol (e.g., Hypertext Transfer Protocol (HTTP)), and/or any other protocol, or set of protocols. In an example embodiment, the network (108) may include, without limitation, a local area network (LAN), a wide area network (WAN) (e.g., the Internet), a mobile network, a virtual network, and/or another suitable public and/or private network capable of supporting communication among two or more of the entities illustrated in Figure 1, or any combination thereof.
[0040] The system (110) is configured to determine real-time fragment distribution for grading the raw materials loaded in each dumper truck (102a and 102b). The system (110) may be embodied in at least one computing device in communication with the network (108). Further, the system (110) may be configured to implement artificial intelligence (AI) techniques for estimating a particle size distribution (PSD) metric for the raw materials being loaded onto each of the dumper trucks (102a; 102b). The PSD metric of the RoM (or the raw materials) indicates the optimality of the blasting process while in turn, it affects the subsequent downstream comminution (or minerals processing).
[0041] In operation, the raw material loading equipment (104) is operated to unload the raw materials to the dumper bed of a dumper truck (e.g., the dumper truck (102a). In one example, the imaging module (106) may be equipped at one or more locations in the raw material loading site. In another example, the imaging module (106) may be mounted at a top portion of the raw material loading equipment (104). This allows the imaging module (106) to be portable along with the raw material loading equipment (104). In another example, the imaging module (106) may be mounted to the dumper trucks (102a; 102b). It is to be noted that the field of view (FOV) of the imaging module (106) is adjusted to capture one or more image frames of the raw materials when loaded onto a dumper truck (e.g., the dumper truck (102a)). In other words, the imaging module (106) captures the image frames of the fragments as the ore particles are loaded into the dumper truck (102a) from a bucket (not shown in figures) of the raw material loading equipment (104).
[0042] In particular, the raw material loading equipment (104) may include a trigger circuit (not shown in figures). The trigger circuit is configured to monitor operations related to the loading of the raw materials contained in the raw material loading equipment (104). The trigger circuit transmits a control signal to the imaging module (106) in response to determining the unloading of the raw materials from the raw material loading equipment (104) into the dumper truck (102a). In other words, the control signal is transmitted to the imaging module (106) at the onset of unloading the raw materials by the raw material loading equipment (104) to the dumper truck (102a). The control signal may be transmitted to the imaging module (104) via one or more communication protocols (e.g., wired or wireless communication protocols).
[0043] The control signal triggers the imaging module (106) to operate for a predetermined time to capture the image frames of the raw materials being loaded into the dumper truck (102a). The predetermined time may be the time from the onset of unloading the raw materials to the dumper truck (102a) till the raw material uniformly settles down completely in the dumper truck (102a) upon unloading. In an embodiment, the imaging module (106) captures the image frames of the raw materials loaded onto the dumper truck (102a) from a specific field of view or a fixed point. In some embodiments, multiple imaging modules (such as the imaging module (106)) may be used for capturing the image frames from multiple field of view. In an embodiment, the imaging module (106) may capture the image frames while the raw materials are contained or loaded in the raw material loading equipment (104) (e.g., the bucket of the shovel).
[0044] Thereafter, the imaging module (106) transmits the image frames of the raw materials to the system (110) in real-time. It is to be noted that the loading of the raw materials to the dumper truck (102a) may be carried out in one instance or in batches depending upon at least a quantity of the raw materials obtained from a blast operation, a payload capacity of the dumper trucks (102a; 102b), and a loading capacity of the raw material loading equipment (104). For example, the payload capacity (or the dumper load (116a)) of the dumper truck (102a) may be 150 tons and the raw material loading equipment (104) may be capable of loading 15 tons (i.e., loading capacity) of raw materials at one instance. In this scenario, the raw material loading equipment (104) is required to operate 10 times to fill 150 tons of the raw material to the dumper truck (102a). Thus, it is understood that the raw materials are loaded onto the dumper truck (102a) in one or more batches (e.g., 10 batches). In this scenario, the control signal is transmitted to the imaging module (106) for capturing the image frames of the raw materials being loaded into the dumper truck (102a) in each batch until the payload capacity of the dumper truck (102a).
[0045] Further, the system (110) receives the image frames of the raw materials being loaded (in batches or a single instance) to the dumper truck (102a). The system (110) may be specifically configured, via executable instructions to perform one or more of the operations described herein. In general, the system (110) is configured to identify a plurality of fragments of the raw materials in the image frames of each batch by implementing artificial intelligence (AI) models. Thereafter, the system (110) may determine a fragment size of each of the plurality of fragments identified in the image frames. Thereafter, the system (110) computes the particle size distribution (PSD) metric of the fragments in each of the dumper trucks (102a; 102b) based on aggregating the fragment size of all the fragments (or the raw materials) loaded in each dumper truck (102a; 102b). Further, the PSD metric of the fragments present in the dumper trucks (102a; 102b) may be at least stored in the database (112) and transmitted to the central interface unit (114) for visualization. The one or more operations performed by the system (110) or any other entity of Figure 1 to estimate the PSD metric are further explained in greater detail.
[0046] The number and arrangement of systems, devices, and/or networks shown in Figure 1 are provided as an example. There may be additional systems, devices, and/or networks; fewer systems, devices, and/or networks; different systems, devices, and/or networks, and/or differently arranged systems, devices, and/or networks than those shown in Figure 1. Furthermore, two or more systems or devices shown in Figure 1 may be implemented within a single system or device, or a single system or device shown in Figure 1 may be implemented as multiple, distributed systems or devices. Additionally, or alternatively, a set of systems (e.g., one or more systems) or a set of devices (e.g., one or more devices) of the environment (100) may perform one or more functions described as being performed by another set of systems or another set of devices of the environment (100).
[0047] Figure 2 illustrates a simplified block diagram representation of a system (200) to determine real-time fragment distribution for grading the raw materials, in accordance with an embodiment of the present disclosure. Examples of the system (200) include but are not limited to, the system (110) as shown in Figure 1. The system (200) includes a computer system (202) and a database (204). The computer system (202) includes at least one processing circuitry (206) for executing instructions, a memory (208), a communication interface (210), and a storage interface (214). The one or more components of the computer system (202) communicate with each other via a bus (212).
[0048] In one embodiment, the database (204) is integrated within the computer system (202) and configured to store the fragment size information, aggregation of the PSD of all the fragments in the dumper trucks from a particular blast, etc. Further, the database (204) may be configured to store at least one artificial intelligence (AI) model (226). The AI models (226) may be trained with training data. The training data may include, but not limited to, historical image data of fragments obtained during the excavation process, the boundary of each fragment in the historical image data, spatial locations of each fragment, and estimated particle size distribution (PSD) values of the fragments in the historical image data. In an embodiment, the historical image data includes the image data of fragments of a particular pixel range (e.g., 1280x720) with different lighting conditions over day and night. In an embodiment, the AI models (226) may include a Mask Region-based Convolutional Neural Network (RCNN) deep learning model. The Mask RCNN deep learning model may be used for object instance segmentation (i.e., detecting objects in an image, delineating their boundaries, and distinguishing between different instances of the same object) which will be explained further in detail. Alternatively, the AI models (226) may include any other deep learning model for performing one or more operations as described herein.
[0049] The computer system (202) may include one or more hard disk drives as the database (204). The storage interface (214) is any component capable of providing the processing circuitry (206) access to the database (204). The storage interface (214) may include, for example, an Advanced Technology Attachment (ATA) adapter, a Serial ATA (SATA) adapter, a Small Computer System Interface (SCSI) adapter, a RAID controller, a SAN adapter, a network adapter, and/or any component providing the processing circuitry (206) with access to the database (204).
[0050] The processing circuitry (206) includes suitable logic and/or interfaces to execute computer-readable instructions. Examples of the processing circuitry (206) include, but are not limited to, an application-specific integrated circuit (ASIC) processor, a reduced instruction set computing (RISC) processor, a complex instruction set computing (CISC) processor, a field-programmable gate array (FPGA), and the like. The memory (208) includes suitable logic, circuitry, and/or interfaces to store a set of computer-readable instructions for performing operations. Examples of the memory (208) include a random-access memory (RAM), a read-only memory (ROM), a removable storage drive, a hard disk drive (HDD), and the like. It will be apparent to a person skilled in the art that the scope of the disclosure is not limited to realizing the memory (208) in the system (200), as described herein. In some embodiments, the memory (208) may be realized in the form of a database server or cloud storage working in conjunction with the system (200), without deviating from the scope of the present disclosure.
[0051] The processing circuitry (206) is communicably coupled to the communication interface (210) such that the processing circuitry (206) is capable of communicating with a remote device (216) such as the imaging module (106), the central interface unit (114), or with any entity connected to the network (108) as shown in Figure 1.
[0052] It is noted that the system (200) as illustrated and hereinafter described is merely illustrative of an apparatus that could benefit from embodiments of the present disclosure and, therefore, should not be taken to limit the scope of the present disclosure. It is noted that the system (200) may include fewer or more components than those depicted in Figure 2.
[0053] In one embodiment, the processing circuitry (206) includes a fragment identification module (218), a fragment size determination module (220), a PSD computing module (222), and a grading module (224).
[0054] The fragment identification module (218) includes a suitable logic and/or interfaces for identifying the fragments of the raw materials in the one or more image frames. As explained above, the raw materials spread out uniformly in the dumper bed when the raw materials are dumped into the dumper truck (e.g., the dumper truck (102a)). In this scenario, the imaging module (106) captures the image frames of the raw materials loaded in the dumper truck (102a) at a single instance or in each batch. Upon receipt of the image frames, the processing circuitry (206) may perform pre-processing operations on the image frames. The pre-processing operations may include enhancing the quality of the data, noise removal, data augmentation, and filtering. Thereafter, the fragment identification module (218) with access to the trained AI models (226) identifies the fragments of the raw materials in the image frames. Further, the AI models (226) may be fine-tuned in real-time based at least on evaluating the fragment identification results of the fragment identification module (218).
[0055] The fragment size determination module (220) includes a suitable logic and/or interfaces to determine the fragment size of each of the plurality of fragments identified in the image frames. In particular, the fragment size determination module (220) defines a fragment region (or bounding boxes) for each of the plurality of fragments of the raw materials in the image frames. In general, the object segmentation is a computer vision task for identifying and delineating the boundaries of the objects within the image frames. To that effect, the fragment size determination module (220) defines the fragment region for each of the fragments based at least on applying one or more object segmentation techniques by the pre-trained AI models (226). In an example, the fragment region may be defined using a coloured mask. The fragment size determination module (220) may apply (or overlay) the coloured mask to the identified fragments in the image frames. In this scenario, the pixels associated with each of the identified fragments (or class of the identified fragments) are highlighted with a unique colour (or distinct colour).
[0056] Thereafter, the fragment size determination module (220) computes the fragment size of each of the plurality of fragments identified in the one or more image frames. In particular, the fragment size determination module (220) applies image processing techniques for identifying at least one fragment dimension element associated with each of the fragments in the image frames. As explained above, coloured mask may be applied (or overlaid) onto the identified fragments. In such cases, the fragment dimension element may include a major axis of the coloured mask. The major axis of an object's (i.e., the identified fragments) shape is the longest straight line that can be drawn within the object. For example, if the segmented object is roughly elliptical in shape, the major axis would be the longest diameter of the ellipse. The fragment size determination module (220) computes the fragment size for each of the fragments based on the analysis of orientation and length of their corresponding major axis (i.e., the fragment dimension element). Additionally, the fragment size determination module (220) may apply a calibration factor for computing the fragment size of each of the fragments. The calibration factor may include at least pixel-to-millimeter (mm) conversion.
[0057] The PSD computing module (222) includes a suitable logic and/or interfaces for computing the particle size distribution (PSD) metric for the plurality of fragments. As explained above, the dumper loads (116a; 116b) correspond to the raw materials being loaded into the dumper trucks (102a; 102b), respectively. Further, the PSD computing module (222) determines the PSD metric for the overall fragments present in each of the dumper loads (116a; 116b). To that effect, the PSD computing module (222) may perform one or more computational operations to aggregate the fragment size of each of the fragments identified in the image frames of each of the dumper loads (116a; 116b), thereby determining the PSD metric for each of the dumper loads (116a; 116b).
[0058] The grading module (224) includes a suitable logic and/or interfaces to estimate at least one quality metric and grading of the fragments. In particular, the grading module (224) aggregates the PSD metric of each of the dumper loads (116a; 116b) to compute the quality metric of the fragments in each of the dumper loads (116a; 116b). This results in the size distribution of the fragments obtained during a particular blast operation. Further, the quality metric indicates the overall blast efficiency of the raw materials obtained in an excavation process (or indicates the optimality of the blast operation). The processing circuitry (206) may utilize the quality metric of the fragments of all the dumper loads (116a; 116b) to optimize or fine-tune the parameters for subsequent blast operations.
[0059] Further, the grading module (224) categorizes the fragments of each of the dumper loads (116a; 116b) based at least on the PSD metric of the fragments in the dumper loads (116a; 116b). It is apparent that categorizing the fragments based on the PSD metric of the dumper loads (116a; 116b) corresponds to grading the fragments. The fragments of the dumper loads (116a; 116b) may be categorized based on the requirements of the downstream operations. For example, certain size fractions may be targeted for optimum crushers for efficient grinding, while others may be critical for flotation or separation.
[0060] Further, the processing circuitry (206) may store the data related to the fragment size of the fragments, the PSD metric of all the dumper loads, the image frames of the fragments, etc., in the database (204) (or the database (112)). In addition, the processing circuitry (206) transmits the particle size distribution (PSD) metric of the dumper loads (116a; 116b) to the central interface unit (114) through the communication interface (210) for visualization. In an embodiment, the central interface unit (114) may include a suitable logic and/or interfaces for grading the fragments of the dumper loads (116a; 116b) based on the PSD metric of the corresponding dumper loads (116a; 116b). The system (200) may use Rest application programming interface (API) for enabling high-speed data flow to indicate the PSD metric of the dumper loads (116a; 116b) on the central interface unit (114) in real-time. The central interface unit (114) may include a web-based visualization system for visualizing the PSD metric of the dumper loads (116a; 116b).
[0061] Figures 3A and 3B illustrate a schematic representation of an image frame (300) depicting fragments of a dumper load in a dumper truck, in accordance with an embodiment of the present disclosure. As shown, the image frame (300) depicts a dumper truck (e.g., the dumper truck (102a) or the dumper truck (102b)) loaded with a dumper load (such as the dumper load (116a) or the dumper load (116b)) using the raw material loading equipment (104). For illustration purposes, the image frame (300) is captured for the dumper truck (102a) with the dumper load (116a). It should be apparent that similar operations for capturing the image frame of the dumper load (116b) of the dumper truck (102b).
[0062] As explained above, the system (200) identifies fragments (see, 302 of Figure 3A) in the dumper load (116a) from the image frame (300) by using the pre-trained AI models (226). In a preferred embodiment, the fragment identification is done every time when the loading is done over dumper truck. Thereafter, the system (200) applies the image processing techniques to determine the fragment size of each of the fragments (302) identified in the image frame (300). The image processing techniques applied to the image frame (300) are exemplarily depicted using a hatching pattern overlaid on the identified fragments (302) (as shown in Figure 3B). The hatching pattern exemplarily represents the coloured mask applied to the identified fragments (302), thereby defining the fragment region of the fragments (302) in the image frame (300) (as shown in Figure 3B). Further, one or more operations related to the identification of the fragments (302), computing the fragment size, etc., are explained with reference to Figure 2, therefore they are not reiterated for the sake of brevity.
[0063] Figure 4A illustrates a graphical representation (400) depicting the particle size distribution (PSD) metric of a dumper load, in accordance with an embodiment of the present disclosure. The graphical representation (400) includes a vertical axis (402) and a horizontal axis (404). The vertical axis (402) represents number of particles (i.e., the fragments) in terms of percentage. The horizontal axis (404) represents a fragment size category. The graphical representation (400) is further depicted to include a plurality of bars for representing the percentage of the fragments in each fragment size category for the dumper load. For example, the percentage of the fragments present in the dumper load for the fragment size category of ‘below 250 (mm)’ is 71% (as shown in Figure 4A). Similarly, aggregating the percentage of the fragments present in each fragment size category provides the PSD metric of the dumper load.
[0064] Figure 4B illustrates a graphical representation (410) depicting a count distribution of the fragments in a dumper load, in accordance with an embodiment of the present disclosure. The graphical representation (420) includes a vertical axis (412) representing the number of particles (i.e., the fragments) in terms of a numerical value and a horizontal axis (414) representing a fragment size category. The graphical representation (410) is further depicted to include a plurality of bars. The length of each bar in the graphical representation (410) indicates the numerical value of the fragments for each fragment size category in the dumper load. For example, the number of the fragments present in the dumper load for the fragment size category of ‘below 250 (mm)’ is ‘671’ (as shown in Figure 4B).
[0065] FIG. 5 illustrates a flow diagram of a computer-implemented method (500) to determine real-time fragment distribution for grading the raw materials in each dumper load, in accordance with an embodiment of the present disclosure. The method (500) depicted in the flow diagram may be executed by, for example, the system (200) or the system (110). Operations of the flow diagram of the method (500), and combinations of the operations in the flow diagram of the method (500), may be implemented by, for example, hardware, firmware, a processor, circuitry, and/or a different device associated with the execution of software that includes one or more computer program instructions. It is noted that the operations of the method (500) can be described and/or practiced by using a system other than these server systems. The method (500) starts at operation (502).
[0066] At operation (502), the method (500) includes receiving, by the processing circuitry (206), one or more image frames comprising raw materials. The raw materials are loaded to the at least one dumper load (116a; 116b).
[0067] At operation (504), the method (500) includes identifying, by the processing circuitry (206), a plurality of fragments of the raw materials in the one or more image frames based at least on at least one artificial intelligence (AI) model.
[0068] At operation (506), the method (500) includes determining, by the processing circuitry (206), a fragment size of each of the plurality of fragments identified in the one or more image frames based at least on implementing image processing techniques.
[0069] At operation (508), the method (500) includes computing, by the processing circuitry (206), a particle size distribution (PSD) metric of the plurality of fragments (302) based at least on aggregating the fragment size of the each of the plurality of fragments identified in the one or more image frames. The particle size distribution (PSD) metric of the plurality of fragments (302) identified in the image frames (300) corresponds to the particle size distribution (PSD) metric of the at least dumper load (116a; 116b). Further, the operations related to real-time fragment distribution determination and grading of the fragments are already explained with reference to Figure 1 to Figures 4A-4B, therefore they are not reiterated, for the sake of brevity.
[0070] Various embodiments of the present invention offer multiple advantages and technical effects. Without limiting the scope of the invention, the systems and methods of the present disclosure determines real-time fragment size distribution for the dumper loads. Further, the fragments are categorized based on the size distribution for grading of the fragments to optimize downstream operations of the fragments.
[0071] While few embodiments of the present invention have been described above, it is to be understood that the invention is not limited to the above embodiments and modifications may be appropriately made thereto within the spirit and scope of the invention.
[0072] While considerable emphasis has been placed herein on the particular features of this invention, it will be appreciated that various modifications can be made and that many changes can be made in the preferred embodiments without departing from the principles of the invention. These and other modifications in the nature of the invention or the preferred embodiments will be apparent to those skilled in the art from the invention herein, whereby it is to be distinctly understood that the foregoing descriptive matter is to be interpreted merely as illustrative of the invention and not as a limitation. , C , Claims:WE CLAIM

1. A system (200) for determining fragment distribution, the system (200) comprising:
a communication interface (210) configured to receive one or more image frames (300) comprising raw materials; and
a processing circuitry (206) communicably coupled to the communication interface (210), the processing circuitry (206) configured to at least:
identify (504) a plurality of fragments (302) of the raw materials in the one or more image frames (300) based, at least in part, on at least one artificial intelligence (AI) model (226),
determine (506) a fragment size of each of the plurality of fragments (302) identified in the one or more image frames (302) based at least on implementing image processing techniques, and
compute (508) a particle size distribution (PSD) metric of the plurality of fragments (302) based at least on aggregating the fragment size of the each of the plurality of fragments (302) identified in the one or more image frames (300).

2. The system (200) as claimed in claim 1, wherein the one or more image frames (300) are captured while the raw materials are loaded in one of: at least one dumper load (116a; 116b) and a raw material loading equipment (104).

3. The system (200) as claimed in claim 1, wherein the particle size distribution (PSD) metric of the plurality of fragments (302) identified in the one or more image frames (300) corresponds to the particle size distribution (PSD) metric of the at least dumper load (116a; 116b).

4. The system (200) as claimed in claim 3, wherein the processing circuitry (206) is further configured to at least:
determine at least one quality metric associated with the plurality of fragments (302) in the at least one dumper load (116a; 116b) based at least on the particle size distribution (PSD) metric of the plurality of fragments (302) in the at least one dumper load (116a; 116b), wherein the at least one quality metric is indicative of an overall blast efficiency of the raw materials obtained in an excavation process.

5. The system (200) as claimed in claim 3, wherein the processing circuitry (206) is further configured to at least:
facilitate grading of the plurality of fragments (302) in the at least one dumper load (116a; 116b) based at least on the particle size distribution (PSD) metric of the plurality of fragments (302) in each of the at least one dumper load (116a; 116b).

6. The system (200) as claimed in claim 1, wherein the processing circuitry (206) is further configured to at least:
define a fragment region for each of the plurality of fragments (302) of the raw materials in the one or more image frames (300) based at least on applying one or more object segmentation techniques by the at least one artificial intelligence (AI) model (226); and
compute the fragment size of each of the plurality of fragments (302) identified in the one or more image frames (300) based at least on identifying at least one fragment dimension element associated with each of the plurality of fragments (302) in the one or more image frames (300) and applying a calibration factor.

7. The system (200) as claimed in claim 1, wherein the at least one artificial intelligence (AI) model (226) is trained with training data, the training data comprising historical image data of fragments obtained during the excavation process, boundary of each fragment in the historical image data, spatial locations of each fragment, and estimated particle size distribution (PSD) values of the fragments in the historical image data.

8. The system (200) as claimed in claim 1, wherein the processing circuitry (206) is further configured to at least:
transmit the particle size distribution (PSD) metric of at least one dumper load (116a; 116b) to a central interface unit (114) through the communication interface (210); and
store data related to the fragment size of each of the plurality of fragments (302), the particle size distribution (PSD) metric associated with each dumper load (116a; 116b) in a database (204) communicably coupled to the processing circuitry (206).

9. The system (200) as claimed in claim 1, further comprising an imaging module (106) mounted to one of a raw material loading equipment (104) and communicably coupled to the processing circuitry (206) through the communication interface (210), the imaging module (106) configured to at least:
receive a control signal from the raw material loading equipment (104) configured to unload the raw materials, wherein the control signal is transmitted to the imaging module (106) at the onset of unloading the raw materials to the at least one dumper load (116a; 116b) in each batch; and
in response to receipt of the control signal, capture the one or more image frames (300) for a predetermined time for each batch of the raw materials being loaded to the at least one dumper load (116a; 116b).

10. A computer-implemented method (500) for determining fragment distribution, comprising:
receiving (502), by a processing circuitry (206), one or more image frames (300) comprising raw materials;
identifying (504), by the processing circuitry (206), a plurality of fragments (302) of the raw materials in the one or more image frames (300) based, at least in part, on at least one artificial intelligence (AI) model (226);
determining (506), by the processing circuitry (206), a fragment size of each of the plurality of fragments (302) identified in the one or more image frames (300) based at least on implementing image processing techniques; and
computing (508), by the processing circuitry (206), a particle size distribution (PSD) metric of the plurality of fragments (302) based at least on aggregating the fragment size of the each of the plurality of fragments (302) identified in the one or more image frames (300).

11. The computer-implemented method (500) as claimed in claim 10, wherein the one or more image frames (300) are captured while the raw materials are loaded in one of: at least one dumper load (116a; 116b) and a raw material loading equipment (104).

12. The computer-implemented method (500) as claimed in claim 10, wherein the particle size distribution (PSD) metric of the plurality of fragments (302) identified in the one or more image frames (300) corresponds to the particle size distribution (PSD) metric of the at least dumper load (116a; 116b).

13. The computer-implemented method (500) as claimed in claim 12, further comprising:
determining, by the processing circuitry (206), at least one quality metric associated with the plurality of fragments (302) in the at least one dumper load (116a; 116b) based at least on the particle size distribution (PSD) metric of the plurality of fragments (302) in the at least one dumper load (116a; 116b), wherein the at least one quality metric is indicative of an overall blast efficiency of the raw materials obtained in an excavation process.

14. The computer-implemented method (500) as claimed in claim 12, further comprising:
facilitating, by the processing circuitry (206), grading of the plurality of fragments (302) in the at least one dumper load (116a; 116b) based at least on the particle size distribution (PSD) metric of the plurality of fragments (302) in each of the at least one dumper load (116a; 116b).

15. The computer-implemented method (500) as claimed in claim 10, further comprising:
defining, by the processing circuitry (206), a fragment region for each of the plurality of fragments (302) of the raw materials in the one or more image frames (300) based at least on applying one or more object segmentation techniques by the at least one artificial intelligence (AI) model (226); and
computing, by the processing circuitry (206), the fragment size of each of the plurality of fragments (302) identified in the one or more image frames (300) based at least on identifying at least one fragment dimension element associated with each of the plurality of fragments (302) in the one or more image frames (300) and applying a calibration factor.

16. The computer-implemented method (500) as claimed in claim 10, wherein the at least one artificial intelligence (AI) model (226) is trained with training data, the training data comprising historical image data of fragments obtained during the excavation process, boundary of each fragment in the historical image data, spatial locations of each fragment, and estimated particle size distribution (PSD) values of the fragments in the historical image data.

17. The computer-implemented method (500) as claimed in claim 10, further comprising:
transmitting, by the processing circuitry (206), the particle size distribution (PSD) metric of at least one dumper load (116a; 116b) to a central interface unit (114) through the communication interface (210); and
storing, by the processing circuitry (206), data related to the fragment size of each of the plurality of fragments (302), the particle size distribution (PSD) metric associated with each dumper load (116a; 116b) in a database (204) communicably coupled to the processing circuitry (206).

18. The computer-implemented method (500) as claimed in claim 10, wherein the one or more image frames (300) are captured by an imaging module (106) communicably coupled to the processing circuitry (206), and wherein the imaging module (106) captures the one or more image frames (300) for a predetermined time for each batch in response to receipt of a control signal from a raw material loading equipment (104) at the onset of unloading the raw materials to the at least one dumper load (116a; 116b).