A method, device and storage medium for identifying overflow state of a flotation froth

Abstract

The embodiment of the present specification provides a flotation froth overflow state identification method, device and storage medium, the method comprises: collecting continuous images in a preset area, the preset area is composed of a flotation tank, an overflow weir and a froth tank; a first deep learning model set in advance obtains a first flotation froth overflow state based on the froth movement state shown in the continuous images, the first flotation froth overflow state includes: flow interruption, tank overflow and others. The technical scheme provided in the present application realizes automatic and timely discovery of abnormal overflow state of froth in the flotation process by using machine vision technology.

Description

TECHNICAL FIELD

[0001] The present document relates to the technical field of flotation, and particularly relates to a flotation froth overflow state recognition method and device and a storage medium. BACKGROUND

[0002] The overflow state of froth in a flotation machine is the most intuitive manifestation of whether the working condition is normal. For example, these states such as flow interruption, tank overflow, too fast or too slow overflow are abnormal states that need to be discovered in time and corrected.

[0003] The existing method sets a distance measuring sensor on the platform of the flotation machine, and determines whether the flow interruption and tank overflow occur through the distance measuring sensor. Therefore, the existing technology needs to additionally set a new device on the platform of the flotation machine to specially detect whether the flow interruption and tank overflow occur in the flotation process.

[0004] However, once the distance measuring sensor fails, replacing and repairing the distance measuring sensor will increase the workload of equipment operation and maintenance, and even cause other unpredictable errors, thereby affecting the efficiency of the flotation process. SUMMARY

[0005] In view of the above analysis, the present application aims to provide a flotation froth overflow state recognition method, device and storage medium to ensure that the abnormal overflow state of froth is automatically and timely discovered in the flotation process.

[0006] In a first aspect, one or more embodiments of the present specification provide a flotation froth overflow state recognition method, comprising:

[0007] collecting continuous images in a preset area, the preset area being composed of a flotation tank, an overflow weir and a froth tank;

[0008] a first pre-set deep learning model obtains a first flotation froth overflow state based on the moving state of froth shown in the continuous images, the first flotation froth overflow state including flow interruption, tank overflow and others.

[0009] Further, when the flotation froth overflow state is others, after the first flotation froth overflow state is obtained, the method further comprises:

[0010] determining the moving speed of froth according to the continuous images;

[0011] determining a second flotation froth overflow state according to the moving speed and a preset speed threshold, the second flotation froth overflow state including too fast overflow, too slow overflow and normal overflow.

[0012] Further, the first deep learning model obtains the first flotation froth overflow state based on the moving state of froth represented in the continuous images, including:

[0013] when the froth continuously moves along the preset direction and the overflow weir cannot be seen on the continuous images, determining that the first flotation froth overflow state is other;

[0014] when the froth does not continuously move along the preset direction and the overflow weir can be seen on the continuous images, determining that the first flotation froth overflow state is broken flow;

[0015] when the froth does not continuously move along the preset direction and the overflow weir cannot be seen on the continuous images, determining that the first flotation froth overflow state is tank overflow.

[0016] Further, the training method of the first deep learning model comprises:

[0017] respectively collecting a first image sequence in a broken flow state, a second image sequence in a tank overflow state, and a third image sequence in an other state;

[0018] training a pre-set deep learning model using the first image sequence, the second image sequence, and the third image sequence as training samples to obtain the first deep learning model.

[0019] Further, the method further comprises:

[0020] randomly taking a single image from the continuous images in the preset area;

[0021] inputting the single image into a pre-set second deep learning model;

[0022] The second deep learning model determines the first flotation froth overflow state based on the single image.

[0023] Further, the second deep learning model determines the first flotation froth overflow state based on the single image, comprising:

[0024] when the overflow weir exists in the single image, determining that the first flotation froth overflow state is broken flow;

[0025] when the overflow weir does not exist in the single image and the froth tank exists, determining that the first flotation froth overflow state is other;

[0026] when the overflow weir and the froth tank do not exist in the single image, determining that the first flotation froth overflow state is tank overflow.

[0027] Further, there is one or more of the preset areas, and each of the preset areas is provided with a shooting device, and the method further comprises:

[0028] determining the shooting range of each shooting device by shooting images;

[0029] for the image corresponding to the shooting range, the shooting device of the foam tank, collect the continuous image in the preset area, and randomly take a single image from the continuous image; input the single image into the second deep learning model set in advance to obtain the first flotation foam overflow state.

[0030] for the image corresponding to the shooting range, the shooting device of the foam tank, collect the continuous image in the preset area, and randomly take a single image from the continuous image; input the single image into the second deep learning model set in advance to obtain the first flotation foam overflow state.

[0031] In a second aspect, the embodiments of the present application provide a flotation foam overflow state recognition device, comprising: a collection module and a data processing module;

[0032] The collection module is used to collect continuous images in a preset area, and the preset area is composed of a flotation tank, an overflow weir and a foam tank.

[0033] The data processing module is used to obtain a first flotation foam overflow state based on the foam movement state displayed in the continuous image, and the first flotation foam overflow state includes: flow interruption, tank overflow and others.

[0034] Further, for the continuous image, the data processing module is used to determine the first flotation foam overflow state as other when the foam moves continuously in a preset direction and the overflow weir does not exist on the continuous image; determine the first flotation foam overflow state as flow interruption when the foam does not move continuously in the preset direction and the overflow weir exists on the continuous image; and determine the first flotation foam overflow state as tank overflow when the foam does not move continuously in the preset direction and the overflow weir does not exist on the continuous image. For the single image, the data processing module is used to determine the first flotation foam overflow state as flow interruption when the overflow weir exists in the single image; determine the first flotation foam overflow state as other when the overflow weir does not exist in the single image but the foam tank exists; and determine the first flotation foam overflow state as tank overflow when the overflow weir and the foam tank do not exist in the single image. When the flotation foam overflow state is other, after obtaining the flotation foam overflow state, the moving speed of the foam is determined according to the continuous image; and the second flotation foam overflow state is determined according to the moving speed and a preset speed threshold.

[0035] In a third aspect, the embodiments of the present application provide a storage medium, comprising:

[0036] for storing computer executable instructions, the computer executable instructions are executed to implement the method of any one of the first aspect.

[0037] Compared with the prior art, the present application can achieve the following technical effects:

[0038] 1. The present application determines the overflow state of the flotation froth through images and software algorithms. By combining with the existing flotation froth image software and hardware system technology, the present application can realize the identification of five overflow states, i.e. overflow chute, flow break, overflow too fast, overflow too slow, and normal overflow, without additional detection equipment.

[0039] 2. The present application uses single images and continuous images to train the second deep learning model and the first deep learning model respectively, and determines the flotation froth overflow state according to the shooting range of the shooting device in combination with the second deep learning model and the first deep learning model, so as to take into account the advantages of less resource occupation using single images for prediction and strong applicability using continuous images for prediction, and further improve the applicability of the method. BRIEF DESCRIPTION OF DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the one or more embodiments of the present specification or the prior art, the drawings needed to be used in the embodiment or prior art description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments described in the present specification, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0041] Figure 1 A flow chart of a flotation froth overflow state identification method provided by one or more embodiments of the present specification;

[0042] Figure 2 A top view of a flotation tank, overflow weir and froth tank provided by one or more embodiments of the present specification. DETAILED DESCRIPTION

[0043] In order to make the person skilled in the art better understand the technical solutions in the one or more embodiments of the present specification, the technical solutions in the one or more embodiments of the present specification will be described clearly and completely below in combination with the drawings in the one or more embodiments of the present specification. Obviously, the described embodiments are only some embodiments of the present specification, not all embodiments. Based on the one or more embodiments of the present specification, all other embodiments obtained by those skilled in the art without creative labor should belong to the protection scope of the present document.

[0044] Different surface visual features can be observed from the top of the flotation machine under different bubble overflow states. When the flow is broken, the froth shakes back and forth in the flotation tank, and both the overflow weir and the froth tank surface of the flotation machine are exposed. When the tank is overflowing, the froth moves in an unsteady direction, and both the overflow weir and the froth tank are covered by froth and ore pulp, and the positions of the overflow weir and the froth tank cannot be distinguished. In addition to the above states, the froth moves in a direction perpendicular to the overflow weir to the outside of the flotation tank, and both the overflow weir and the froth tank are covered by froth and ore pulp, but the positions of the overflow weir and the froth tank can be roughly distinguished.

[0045] However, the above visual features are based on the observation of the human eye, but for a shooting device fixed on the flotation machine, its shooting range is fixed and cannot be adjusted according to the actual situation, so the shooting device may shoot an incomplete overflow weir and froth tank, and it is difficult to distinguish the bubble overflow, broken flow and other overflow states based on the incomplete overflow weir and froth tank.

[0046] Based on the above scenario, the embodiments of the present specification provide a flotation froth overflow state recognition method, as shown in the following formula: Figure 1 The method comprises the following steps:

[0047] Step 1, collect continuous images in a preset area.

[0048] In the embodiments of the present application, the preset area is composed of a flotation tank, an overflow weir and a froth tank. At the same time, a plurality of shooting devices or one shooting device are arranged in the preset area. When there are a plurality of shooting devices, each shooting device collects continuous images, and respectively determines the flotation froth overflow state of the area responsible for by the corresponding shooting device according to the collected continuous images. The shooting device includes an image acquisition device and a video acquisition device. If it is an image acquisition device, it directly collects continuous images in the preset area. If it is a video acquisition device, it acquires a video in the preset area and decodes to collect continuous images in the preset area.

[0049] Step 2, a first deep learning model pre-set based on the froth movement state shown in the continuous images to obtain a first flotation froth overflow state.

[0050] In the embodiments of the present application, in order to reduce the training samples, the deep learning model selects a pre-trained model such as VGG-16, ResNet-50, MobileNet-V2, etc. The training process of the first deep learning model is as follows: a first image sequence under a broken flow state, a second image sequence under a tank overflowing state, and a third image sequence under other states are collected. The first image sequence, the second image sequence and the third image sequence are used as training samples to train the pre-trained model to obtain the first deep learning model.

[0051] In the embodiments of the present application, the first flotation froth overflow state includes: flow interruption, tank overflow and others. Based on the different movement states of the froth in the flow interruption and tank overflow states, the criterion for the first flotation froth overflow state is:

[0052] When the froth continuously moves along the preset direction and the overflow weir does not exist on the continuous image, it is determined that the first flotation froth overflow state is others;

[0053] When the froth does not continuously move along the preset direction and the overflow weir exists on the continuous image, it is determined that the first flotation froth overflow state is flow interruption;

[0054] When the froth does not continuously move along the preset direction and the overflow weir does not exist on the continuous image, it is determined that the first flotation froth overflow state is tank overflow.

[0055] Wherein, the preset direction and whether the overflow weir exists on the continuous image are information obtained by the first deep learning model through training, without manual specification.

[0056] Specifically, the shooting device is fixedly arranged on the flotation machine, that is, in the corresponding shooting direction, the flow direction of the froth in other states is unchanged, so the deep learning model can extract the corresponding features based on the flow direction of the froth in the state, and convert the features into the preset flow direction.

[0057] In the embodiments of the present application, based on the continuous image, in addition to being able to distinguish between flow interruption and tank overflow, the flotation froth overflow in other states can be classified more finely.

[0058] Specifically, according to the continuous image, the moving speed of the froth is determined. According to the moving speed and the preset speed threshold, the second flotation froth overflow state is determined. Wherein, the second flotation froth overflow state includes: overflow too fast, overflow too slow and overflow normal. Through the above-mentioned manner, the detection personnel can understand the flow situation of the froth in the flotation machine, so as to predict the flotation machine that may appear flow interruption or tank overflow.

[0059] In the embodiments of the present application, when the flotation froth overflow state is overflow too fast, it indicates that the flotation process may appear tank overflow, at this time, the aeration amount needs to be reduced or the liquid level needs to be lowered to avoid tank overflow. When the flotation froth overflow state is overflow too slow, it indicates that the flotation process may appear flow interruption, at this time, the aeration amount needs to be increased and the liquid level needs to be raised to avoid flow interruption. When the flotation froth overflow state is overflow normal, at this time, the aeration amount and the liquid level remain unchanged. It can be known that the technical scheme provided by the present application can help the staff to judge in advance whether flow interruption and tank overflow will occur, so as to take measures in advance, thereby avoiding flow interruption or tank overflow.

[0060] The software algorithm processing continuous images will occupy a large amount of system resources, thereby increasing the data processing burden. Therefore, in the embodiment of the present application, the overflow state of the flotation froth can also be determined by using a single image, and the specific process is as follows:

[0061] A single image is randomly obtained from the continuous images in the preset area. The single image is input into a second pre-set deep learning model. The second deep learning model determines the first flotation froth overflow state based on the single image. Wherein, the second deep learning model is obtained by training a pre-trained model, and the pre-trained model can be selected from VGG-16, ResNet-50, MobileNet-V2 and the like.

[0062] Wherein, the specific process of randomly obtaining a single image from the continuous images in the preset area is as follows:

[0063] The continuous images in the preset area are continuously collected by the shooting device, and the continuous images are stored in the preset queue. Then, a single image is randomly extracted from the queue.

[0064] The criterion of the flotation froth overflow state corresponding to the second deep learning model is as follows:

[0065] When the single image contains an overflow weir, the first flotation froth overflow state is determined to be broken flow;

[0066] When the single image contains only a froth tank, the first flotation froth overflow state is determined to be other;

[0067] When the single image does not contain an overflow weir and a froth tank, the first flotation froth overflow state is determined to be a froth tank.

[0068] Wherein, whether the single image contains an overflow weir or a froth tank is information obtained by the second deep learning model through training, without manual specification.

[0069] Based on the above method, when a flotation site is provided with multiple shooting devices, the process of determining the flotation froth overflow state is as follows:

[0070] The shooting range of each shooting device is determined by shooting images;

[0071] For the shooting device whose corresponding image does not contain a froth tank, continuous images in the preset area are collected; the continuous images are input into a first pre-set deep learning model to obtain a first flotation froth overflow state;

[0072] For the shooting device whose corresponding image contains a froth tank, a single image is randomly obtained from the continuous images after collecting the continuous images in the preset area; the single image is input into a second pre-set deep learning model to obtain a first flotation froth overflow state.

[0073] Specifically, when multiple devices are installed on the flotation machine platform, the relative positions cannot be guaranteed to be consistent due to external factors or installation accuracy problems, and the shooting ranges of the shooting devices may not be the same. Some can shoot the flotation tank, overflow weir and froth tank, and some can only shoot the flotation tank and overflow weir. The structures of the flotation tank, overflow weir and froth tank are as shown in Figure 2 Figure 2 is a top view of the flotation tank 1, overflow weir 2 and froth tank 3. For the case where the shooting range contains the flotation tank, overflow weir and froth tank, the overflow state of the froth is identified according to the criterion corresponding to the second deep learning model. For the case where the shooting range only contains the flotation tank and the overflow weir, the overflow state of the froth needs to be identified according to the criterion corresponding to the first deep learning model.

[0074] Similarly, for the case where multiple shooting devices are arranged on one flotation machine, the shooting ranges of the shooting devices may not be the same. Similarly, for the image in which the shooting range contains the flotation tank, overflow weir and froth tank, the overflow state of the froth is identified according to the criterion corresponding to the second deep learning model. For the image in which the shooting range only contains the flotation tank and the overflow weir, the overflow state of the froth needs to be identified according to the criterion corresponding to the first deep learning model.

[0075] In the above manner, the problem of poor model applicability caused by different shooting ranges of multiple shooting devices can be avoided. It should be noted that for the case where the first deep learning model is applicable to some shooting devices and the second deep learning model is applicable to some shooting devices, all shooting devices can be caused to use the first deep learning model.

[0076] The embodiment of the present application provides a flotation froth overflow state identification device, which comprises a collection module and a data processing module.

[0077] The collection module is used for collecting continuous images in a preset area, and the preset area is composed of a flotation tank, an overflow weir and a froth tank.

[0078] The data processing module is used for obtaining first and second flotation froth overflow states based on the moving state of the froth shown in the continuous images or the image features shown in the images randomly taken from the continuous images. The first flotation froth overflow state comprises cut-off, tank overflow and others, and the second flotation froth overflow state comprises overflow too fast, overflow too slow and normal overflow.

[0079] ​In the embodiment of the present application, for the continuous images, the data processing module is configured to determine the first flotation froth overflow state as other when the froth continuously moves along the preset direction and the overflow weir does not exist on the continuous images; determine the first flotation froth overflow state as flow interruption when the froth does not continuously move along the preset direction and the overflow weir exists on the continuous images; and determine the first flotation froth overflow state as froth tank overflow when the froth does not continuously move along the preset direction and the overflow weir does not exist on the continuous images. For the single image randomly taken from the continuous images, the data processing module is configured to determine the first flotation froth overflow state as flow interruption when the overflow weir exists in the single image; determine the first flotation froth overflow state as other when the overflow weir does not exist in the single image but the froth tank exists; and determine the first flotation froth overflow state as froth tank overflow when the overflow weir and the froth tank do not exist in the single image. When the flotation froth overflow state is other, after obtaining the flotation froth overflow state, the data processing module is configured to determine the moving speed of the froth according to the continuous images; and determine the second flotation froth overflow state according to the moving speed and a preset speed threshold.

[0080] The embodiment of the present application provides a storage medium, characterized in comprising:

[0081] The computer executable instructions are used for storing the method of any one of the above embodiments.

[0082] The above describes specific embodiments of the present specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be executed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are possible or can be advantageous.

[0083] In the 1930s, it was clear to distinguish whether an improvement in a technology was in hardware (e.g., improvement in circuit structure of diodes, transistors, switches, etc.) or in software (e.g., improvement in method flow). However, as technology has evolved, many improvements in method flow today can be considered as direct improvements in hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that an improvement in a method flow cannot be implemented by a hardware entity module. For example, a programmable logic device (PLD) (e.g., a field programmable gate array (FPGA)) is an integrated circuit whose logic function is determined by user programming of the device. A digital system is "integrated" on a PLD by the designer programming it, rather than by asking a chip manufacturer to design and fabricate a custom integrated circuit chip. Moreover, instead of manually fabricating an integrated circuit chip, this programming is now mostly implemented by "logic compiler" software, which is similar to software compilers used in program development, and the original code to be compiled is written in a specific programming language, called a hardware description language (HDL), of which there are many, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc., the most commonly used being VHDL (Very-High-Speed Integrated Circuit Hardware Description Language) and Verilog. It should be clear to those skilled in the art that, by simply logically programming a method flow in one of the above hardware description languages and programming it into an integrated circuit, a hardware circuit implementing the logical method flow can be easily obtained.

[0084] The controller can be implemented in any suitable way, for example, the controller can take the form of a microprocessor or processor and a computer readable medium storing computer readable program code, such as software or firmware, executable by the (micro)processor, logic gates, switches, an application specific integrated circuit (ASIC), a programmable logic controller and an embedded microcontroller, examples of which include but are not limited to the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20 and Silicone Labs C8051F320, the memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also know that, in addition to implementing the controller purely in terms of computer readable program code, it is possible to implement the controller to perform the same functions in terms of logic gates, switches, application specific integrated circuits, programmable logic controllers and embedded microcontrollers, by suitably programming the logic gates, switches, application specific integrated circuits, programmable logic controllers and embedded microcontrollers. The controller can therefore be considered a hardware component, and the means for performing the various functions comprised within it can be considered structures within the hardware component. Alternatively, or even additionally, the means for performing the various functions can be considered both software modules which implement the methods and structures within the hardware component.

[0085] The systems, apparatuses, modules or units illustrated by the above embodiments can be specifically implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0086] For the sake of description, the above apparatuses are described in functional division and are described respectively. Of course, the functions of the units can be implemented in the same or multiple software and / or hardware when implementing the embodiments of the present specification.

[0087] Those skilled in the art will understand that one or more embodiments of the present specification can be provided as a method, a system or a computer program product. Therefore, one or more embodiments of the present specification can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present specification can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0088] The specification is presented with reference to flow diagrams and / or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the specification. It will be understood that each block of the flow diagrams and / or block diagrams, and combinations of blocks in the flow diagrams and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processing element or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flow diagrams and / or block diagrams block or blocks. Figure 1 The flow diagrams and / or block diagrams in the specification can present a method, apparatus or computer program product according to embodiments of the specification. Flow diagrams and / or block diagrams can also present a method, apparatus or computer program product to achieve functions specified in flow diagrams and / or block diagrams block or blocks. Figure 1 The flow diagrams and / or block diagrams in the specification can present a method, apparatus or computer program product according to embodiments of the specification. Flow diagrams and / or block diagrams can also present a method, apparatus or computer program product to achieve functions specified in flow diagrams and / or block diagrams block or blocks.

[0089] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the flow diagrams and / or block diagrams block or blocks. Figure 1 The flow diagrams and / or block diagrams in the specification can present a method, apparatus or computer program product according to embodiments of the specification. Flow diagrams and / or block diagrams can also present a method, apparatus or computer program product to achieve functions specified in flow diagrams and / or block diagrams block or blocks. Figure 1 The flow diagrams and / or block diagrams in the specification can present a method, apparatus or computer program product according to embodiments of the specification. Flow diagrams and / or block diagrams can also present a method, apparatus or computer program product to achieve functions specified in flow diagrams and / or block diagrams block or blocks.

[0090] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the flow diagrams and / or block diagrams block or blocks. Figure 1 The flow diagrams and / or block diagrams in the specification can present a method, apparatus or computer program product according to embodiments of the specification. Flow diagrams and / or block diagrams can also present a method, apparatus or computer program product to achieve functions specified in flow diagrams and / or block diagrams block or blocks. Figure 1 The flow diagrams and / or block diagrams in the specification can present a method, apparatus or computer program product according to embodiments of the specification. Flow diagrams and / or block diagrams can also present a method, apparatus or computer program product to achieve functions specified in flow diagrams and / or block diagrams block or blocks.

[0091] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0092] The memory can include non-persistent memory and / or storage mechanisms such as, for example, random access memory (RAM), non-volatile memory (NVM), and / or a persistent memory such as, for example, read-only memory (ROM) or flash memory. The memory is an example of computer-readable media.

[0093] Computer-readable media includes permanent and non-permanent, movable and non-movable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible to a computing device. According to the definition herein, computer-readable media does not include transitory media such as modulated data signals and carriers.

[0094] It should also be noted that the terms "comprising", "containing", or any other variant thereof are intended to cover non-exclusive inclusion, such that processes, methods, articles or devices that include a series of elements not only include those elements, but also include other elements not explicitly listed, or inherent to such processes, methods, articles or devices. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of additional identical elements in the process, method, article or device that includes the element.

[0095] One or more embodiments of the specification can be described in the general context of computer-executable instructions being executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. One or more embodiments of the specification can also be practiced in a distributed computing environment, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media, including storage devices.

[0096] Various embodiments in the specification are described in a progressive manner, and the same or similar parts between various embodiments can be referred to each other, and each embodiment focuses on the difference from other embodiments. In particular, for system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and the relevant parts can be referred to the part of the method embodiment.

[0097] The above merely provides the example of the present document and is not intended to limit the present document. For those skilled in the art, the present document can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present document shall be included in the scope of claims of the present document.

Claims

1. A method for identifying froth overflows in a flotation process, characterized in that, The method comprises: collecting continuous images in a preset area, the preset area being composed of a flotation tank, an overflow weir and a froth tank; a first deep learning model is pre-set to obtain a first flotation froth overflow state based on a froth movement state shown in the continuous images, the first flotation froth overflow state comprising: flow interruption, tank overflow and others; the first deep learning model obtains a first flotation froth overflow state based on a froth movement state represented in the continuous images, comprising: when the froth continuously moves in a preset direction and the continuous image does not exist the overflow weir, it is determined that the first flotation froth overflow state is other; when the froth does not continuously move in a preset direction and the continuous image exists the overflow weir, it is determined that the first flotation froth overflow state is flow interruption; when the froth does not continuously move in a preset direction and the continuous image does not exist the overflow weir, it is determined that the first flotation froth overflow state is tank overflow.

2. The method of claim 1, wherein when the first flotation froth overflow state is other, the method further comprises, after obtaining the first flotation froth overflow state: determining the moving speed of the froth according to the continuous images; determining a second flotation froth overflow state according to the moving speed and a preset speed threshold, the second flotation froth overflow state comprising: overflow too fast, overflow too slow and overflow normal.

3. The method of claim 1, wherein the training method of the first deep learning model comprises: collecting a first image sequence in a flow interruption state, a second image sequence in a tank overflow state and a third image sequence in an other state respectively; training a pre-set deep learning model to obtain the first deep learning model using the first image sequence, the second image sequence and the third image sequence as training samples. The method further comprises: randomly taking a single image from the continuous images in the preset area; 4. The method of claim 1, wherein, inputting the single image into a pre-set second deep learning model; the second deep learning model determines the first flotation froth overflow state based on the single image.

5. The method of claim 4, wherein the second deep learning model determines the first flotation froth overflow state based on the single image, comprising: when the single image exists the overflow weir, it is determined that the first flotation froth overflow state is flow interruption; when the single image does not exist the overflow weir but exists the froth tank, it is determined that the first flotation froth overflow state is other; when the single image does not exist the overflow weir and the froth tank, it is determined that the first flotation froth overflow state is tank overflow. There is one or more of the preset areas, and each of the preset areas is provided with a shooting device, the method further comprising: determining the shooting range of each shooting device by shooting images; for the shooting device whose corresponding image does not exist the froth tank, collecting the continuous images in the preset area; inputting the continuous images into the pre-set first deep learning model to obtain the first flotation froth overflow state; 6. The method of claim 4, wherein, ​ ​ ​ For the image corresponding to the shooting range, there is a shooting device of the froth tank, after collecting continuous images in the preset area, a single image is randomly taken from the continuous images; the single image is input into a second deep learning model set in advance to obtain the first flotation froth overflow state.

7. A device for recognizing the state of overflow of a flotation froth, characterized in that Comprise: a collection module and a data processing module; The collection module is used to collect continuous images in a preset area, and the preset area is composed of a flotation tank, an overflow weir and a froth tank; The data processing module is used to obtain a first flotation froth overflow state based on the froth movement state shown in the continuous images or the image features shown in the single image randomly taken from the continuous images, the first flotation froth overflow state comprising: flow interruption, tank overflow and others; For continuous images, the data processing module is used to determine the first flotation froth overflow state as other when the froth continuously moves in a preset direction and the overflow weir does not exist on the continuous image; determine the first flotation froth overflow state as flow interruption when the froth does not continuously move in a preset direction and the overflow weir exists on the continuous image; determine the first flotation froth overflow state as tank overflow when the froth does not continuously move in a preset direction and the overflow weir does not exist on the continuous image.

8. The apparatus of claim 7, wherein, For a single image, the data processing module is used to determine the first flotation froth overflow state as flow interruption when the overflow weir exists in the single image; determine the first flotation froth overflow state as other when the overflow weir does not exist in the single image but the froth tank exists; determine the first flotation froth overflow state as tank overflow when the overflow weir and the froth tank do not exist in the single image; When the first flotation froth overflow state is other, after obtaining the first flotation froth overflow state, the moving speed of the froth is determined according to the continuous images; the second flotation froth overflow state is determined according to the moving speed and the preset speed threshold.

9. A storage medium, comprising: computer executable instructions for storing, the computer executable instructions being executed to implement the method of any one of claims 1-6.

Images