KR20230149262A - Method, server and computer program for detecting moving object in image based on wavelength information - Google Patents

Abstract

A method for detecting a moving object in an image based on wavelength information corresponding to color according to various embodiments of the present invention for realizing the above-described tasks is disclosed. The method is performed in one or more processors of a computing device, comprising: acquiring image data; generating a plurality of wavelength information corresponding to the image data based on a plurality of color information included in the image data; and detecting a moving object based on the plurality of wavelength information.

Description

Method for detecting moving objects in images based on wavelength information, server and computer program {METHOD, SERVER AND COMPUTER PROGRAM FOR DETECTING MOVING OBJECT IN IMAGE BASED ON WAVELENGTH INFORMATION}

Various embodiments of the present invention relate to a method for detecting moving objects. More specifically, a method, server, and method for distinguishing objects based on wavelength information corresponding to color and detecting and tracking moving objects in real time based on this. It's about computer programs.

Today, with the development of artificial intelligence, many conveniences are provided by using computers to handle tasks that should be done by humans. In particular, object detection is a field of research that allows computers to analyze and interpret visual information, which is where humans receive the most information. This object detection is used in video surveillance, facial recognition, robot control, IoT, autonomous driving, manufacturing, security, etc., making it an essential core technology across industries and being used in a variety of fields.

For example, object detection technology can be used in the field of autonomous vehicles. For autonomous driving, a computer must be able to detect and recognize surrounding objects on its own. For example, object detection technology can be applied to autonomous vehicles that require interaction with the surrounding environment, such as slowing down when there is a stop sign and starting again when the green light comes on.

Additionally, for example, object detection technology can be utilized for efficient resource management in the security field. In general, CCTV requires a huge amount of memory because it records continuously, but by combining it with object detection technology and recording when a specific object is detected, memory can be used and managed more efficiently. Object detection technology is actively used in various fields other than the examples described above (for example, medical care or robot control, etc.).

Meanwhile, in general, objects can be detected in an image through an object detection method using a probability distribution corresponding to the similarity of the color histogram. In this way, the object detection method using the color information of the object has the advantage of being simple to implement and robust to object shape transformation.

However, the object detection method using color information of objects may not be able to continuously detect and track each object when overlap between moving objects of similar colors occurs in a situation where multiple objects are mixed. In addition, the object detection method using the color information of the object loses the shape information of the object, so it is easy to miss the object when there is a color similar to the object in the background, and is vulnerable to changes in the size of the object. In particular, when the image acquisition module that acquires images is installed in a moving device (e.g., a vehicle), the size of the target or object to be identified in the image continuously changes, so errors in object detection are more likely to occur in this situation. There are concerns that it will grow.

Korean Patent Publication No. 10-2010-0118162 (2010.11.05.)

The problem to be solved by the present invention was developed in response to the above-mentioned background technology, and provides a method of tracking moving objects in real time according to the correlation between wavelength and absorption rate based on color and color models such as RGB, HIS, and YCrCb. This is to do so.

The problems to be solved by the present invention are not limited to the problems mentioned above, and other problems not mentioned can be clearly understood by those skilled in the art from the description below.

To solve the above-described problem, a method for detecting a moving object in an image based on wavelength information corresponding to color according to an embodiment of the present invention is disclosed. The method is performed in one or more processors of a computing device, comprising: acquiring image data; generating a plurality of wavelength information corresponding to the image data based on a plurality of color information included in the image data; and detecting a moving object based on the plurality of wavelength information.

In an alternative embodiment, detecting a moving object based on the plurality of wavelength information includes identifying a plurality of objects included in the image data based on the plurality of wavelength information, It may include distinguishing each of the plurality of objects and predicting a movement path of each of the plurality of divided objects based on changes in the plurality of wavelength information to generate movement path prediction information.

In an alternative embodiment, the step of detecting a moving object based on the plurality of wavelength information includes identifying at least two or more wavelength information related to overlap among the plurality of wavelength information, and detecting the overlap in the image data. It may include generating corrected image data by removing non-corresponding objects and detecting a moving object based on wavelength information included in the corrected image data.

In an alternative embodiment, the step of detecting a moving object based on the wavelength information included in the corrected image data includes calculating a probability value based on the wavelength information included in the corrected image data, and based on the probability value. It includes the step of detecting a moving object, wherein the probability value is a point that changes according to the movement of the object included in the image data, and may be characterized as being a standard for distinguishing the boundary of the wavelength.

In an alternative embodiment, calculating the probability value includes identifying an intersection area of first and second wavelength information related to overlap and calculating the probability value based on the center point of the intersection area. can do.

In an alternative embodiment, the step of detecting a moving object based on the probability value includes identifying which direction of the first wavelength information direction and the second wavelength information direction the location of each pixel is located based on the probability value. , calculating a distance from the probability value for each pixel; And it may include detecting a moving object based on the location and distance identified corresponding to each pixel.

In an alternative embodiment, calculating the probability value includes evaluating the similarity of the first wavelength information and the second wavelength information related to overlap, wherein the similarity of the first wavelength information and the second wavelength information is greater than or equal to a predetermined similarity degree. In this case, assigning random wavelength information different from each other to each of the first wavelength information and the second wavelength information, identifying an intersection area between the random wavelength information assigned to each object, and based on the center point of the intersection area It may include calculating the probability value.

In an alternative embodiment, the method further includes performing preprocessing corresponding to the image data, wherein the preprocessing may include preprocessing related to a morphology operation to remove noise due to illuminance changes and movement. there is.

According to another embodiment of the present invention, a server for performing a method of detecting a moving object in an image based on wavelength information corresponding to color is disclosed. The server includes a memory that stores one or more instructions and a processor that executes one or more instructions stored in the memory, and the processor executes the one or more instructions, thereby creating an image in an image based on wavelength information corresponding to the color described above. A method for detecting moving objects can be performed.

According to another embodiment of the present invention, a computer program stored in a computer-readable recording medium is disclosed. The computer program can be combined with a hardware computer to perform a method of detecting a moving object in an image based on wavelength information corresponding to color.

Other specific details of the invention are included in the detailed description and drawings.

According to various embodiments of the present invention, a robust object detection and tracking method is provided by detecting moving objects in an image using color wavelength information, even in situations where multiple objects are mixed and overlap between moving objects of similar colors occurs. can be provided.

The effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below.

Figure 1 is an exemplary diagram schematically showing a system for implementing a method for detecting a moving object in an image based on wavelength information corresponding to color related to an embodiment of the present invention.
Figure 2 is a hardware configuration diagram of a server that performs an embodiment of the present invention and a method of detecting a moving object in an image based on wavelength information corresponding to color.
Figure 3 shows a flowchart illustrating a method for detecting a moving object in an image based on wavelength information corresponding to color related to an embodiment of the present invention.
Figure 4 is a diagram showing that the wavelength values are different for each color of the present invention.
Figure 5 is an exemplary diagram illustrating segmentation of image data including objects of various colors related to an embodiment of the present invention.
Figure 6 shows a flowchart illustrating an example of a method for correcting image data related to an embodiment of the present invention.
Figure 7 is an example diagram for explaining the process for calculating the probability value of the present invention.
Figure 8 is an exemplary diagram illustrating a process for generating a model for object detection related to an embodiment of the present invention.
Figure 9 is an example diagram to explain the moving object detection process of the present invention.

Various embodiments are now described with reference to the drawings. In this specification, various descriptions are presented to provide an understanding of the invention. However, it is clear that these embodiments may be practiced without these specific descriptions.

As used herein, the terms “component,” “module,” “system,” and the like refer to a computer-related entity, hardware, firmware, software, a combination of software and hardware, or an implementation of software. For example, a component may be, but is not limited to, a process running on a processor, a processor, an object, a thread of execution, a program, and/or a computer. For example, both an application running on a computing device and the computing device can be a component. One or more components may reside within a processor and/or thread of execution. A component may be localized within one computer. A component may be distributed between two or more computers. Additionally, these components can execute from various computer-readable media having various data structures stored thereon. Components can transmit signals, for example, with one or more data packets (e.g., data and/or signals from one component interacting with other components in a local system, a distributed system, to other systems and over a network such as the Internet). Depending on the data being transmitted, they may communicate through local and/or remote processes.

Additionally, the term “or” is intended to mean an inclusive “or” and not an exclusive “or.” That is, unless otherwise specified or clear from context, “X utilizes A or B” is intended to mean one of the natural implicit substitutions. That is, either X uses A; X uses B; Or, if X uses both A and B, “X uses A or B” can apply to either of these cases. Additionally, the term “and/or” as used herein should be understood to refer to and include all possible combinations of one or more of the related listed items.

Additionally, the terms “comprise” and/or “comprising” should be understood to mean that the corresponding feature and/or element is present. However, the terms “comprise” and/or “comprising” should be understood as not excluding the presence or addition of one or more other features, elements and/or groups thereof. Additionally, unless otherwise specified or the context is clear to indicate a singular form, the singular terms herein and in the claims should generally be construed to mean “one or more.”

Those skilled in the art will additionally recognize that the various illustrative logical blocks, components, modules, circuits, means, logic, and algorithm steps described in connection with the embodiments disclosed herein may be implemented using electronic hardware, computer software, or a combination of both. It must be recognized that it can be implemented with To clearly illustrate the interchangeability of hardware and software, various illustrative components, blocks, configurations, means, logics, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented in hardware or software will depend on the specific application and design constraints imposed on the overall system. A skilled technician can implement the described functionality in a variety of ways for each specific application. However, such implementation decisions should not be construed as departing from the scope of the present invention.

The description of the presented embodiments is provided to enable anyone skilled in the art to use or practice the present invention. Various modifications to these embodiments will be apparent to those skilled in the art. The general principles defined herein may be applied to other embodiments without departing from the scope of the invention. Therefore, the present invention is not limited to the embodiments presented herein. The present invention is to be interpreted in the broadest scope consistent with the principles and novel features presented herein.

In this specification, a computer refers to all types of hardware devices including at least one processor, and depending on the embodiment, it may be understood as encompassing software configurations that operate on the hardware device. For example, a computer can be understood to include, but is not limited to, a smartphone, tablet PC, desktop, laptop, and user clients and applications running on each device.

Hereinafter, embodiments of the present invention will be described in detail with reference to the attached drawings.

Each step described in this specification is described as being performed by a computer, but the subject of each step is not limited thereto, and depending on the embodiment, at least part of each step may be performed in a different device.

Figure 1 is an exemplary diagram schematically showing a system for implementing a method for detecting a moving object in an image based on wavelength information corresponding to color related to an embodiment of the present invention.

As shown in FIG. 1, a system according to embodiments of the present invention may include a server 100, a user terminal 200, an external server 300, and a network 400. The components shown in FIG. 1 are exemplary, and additional components may exist or some of the components shown in FIG. 1 may be omitted. The server 100, the external server 300, and the user terminal 200 according to embodiments of the present invention can mutually transmit and receive data for the system according to embodiments of the present invention through the network 400. there is.

The network 400 according to embodiments of the present invention includes Public Switched Telephone Network (PSTN), x Digital Subscriber Line (xDSL), Rate Adaptive DSL (RADSL), Multi Rate DSL (MDSL), and Very Various wired communication systems such as High Speed DSL), Universal Asymmetric DSL (UADSL), High Bit Rate DSL (HDSL), and Local Area Network (LAN) can be used.

In addition, the network 400 presented herein includes Code Division Multi Access (CDMA), Time Division Multi Access (TDMA), Frequency Division Multi Access (FDMA), Orthogonal Frequency Division Multi Access (OFDMA), and Single Carrier-FDMA (SC-FDMA). A variety of wireless communication systems may be used, such as FDMA) and other systems.

The network 400 according to embodiments of the present invention can be configured regardless of the communication mode, such as wired or wireless, and can be configured in various communication modes such as a personal area network (PAN) and a wide area network (WAN). It may consist of a communication network. In addition, the network 400 may be the known World Wide Web (WWW), and may use wireless transmission technology used for short-distance communication, such as Infrared Data Association (IrDA) or Bluetooth. . The techniques described herein can be used in the networks mentioned above, as well as other networks.

According to an embodiment of the present invention, the server 100 (hereinafter referred to as 'server 100'), which performs a method of detecting a moving object in an image based on wavelength information corresponding to color, detects a moving object in the image data. You can. Specifically, the server 100 can acquire image data in real time and automatically detect moving objects in the real-time image data.

According to one embodiment, the server 100 may detect an object in image data based on color. Specifically, the server 100 can generate a plurality of wavelength information corresponding to the image data based on a plurality of color information included in the image data, and can detect a moving object based on the wavelength information. Each color has a different wavelength, and the server 100 can detect moving objects based on the wavelengths of these colors. For example, red may be 750 nm, purple may be 410 nm, and the wavelength value may be different for each color. Using the wavelength of visible light of the present invention, moving objects in an image can be detected.

The server 100 may identify and distinguish each of a plurality of objects in the image data based on each wavelength corresponding to each color. That is, the server 100 can perform segmentation for each object in the image based on color.

Generally, in the field of object detection technology, a method of detecting objects using the similarity of color histograms is mainly used. Technology for detecting objects based on color (i.e., color factor-based) obtains the probability distribution of the colors constituting the target object from the image, then projects this back onto the input image to obtain the probability distribution of the object corresponding to each pixel color. Calculate and use its accumulated value to detect the target object. As a specific example, the Meanshift method of object detection is being used, which detects target objects by searching for the point where the probability density is locally maximized within the probability image. This color-based search method has the advantage of being simple to implement, fast, and robust to changes in the shape of the object, but because it loses the shape information of the object, there is a color similar to the object in the background, or there is overlap between objects of similar colors. When this phenomenon occurs, it is difficult to detect the target object and there is a problem in that it is vulnerable to changes in the size of the target object. In particular, in the case of images acquired from moving devices (e.g., vehicles for autonomous driving), the size of the target or object to be identified within the image continuously changes, making object detection more difficult in such situations, thereby reducing accuracy and reliability. may deteriorate. Additionally, in the conventional color-based object detection technology, it is not easy to determine the size of the search radius, which may reduce the accuracy of object detection. For example, when detecting and tracking a moving object in an image, the size or shape of the search window must be appropriately changed according to changes in the size and shape of the target object. If the search window does not change appropriately, object detection and tracking may be impaired. It has a big impact.

The present invention is intended to complement the problems in the conventional color-based object detection technology. Rather than detecting a moving object based on the color value of the color itself (e.g., color histogram), the present invention detects a moving object based on the color wave length value. It is characterized by detecting moving objects based on

According to an embodiment, the server 100 segments each of a plurality of objects in the image based on the wavelength value of the color, and predicts the movement path of each object based on the segmentation result and the change in each wavelength, or Overlap between objects can be identified. When an overlap between each object is identified, the server 100 identifies whether the objects related to the overlap are objects corresponding to different wavelength values, and when the difference in the wavelength values of each object is greater than a certain level, the server 100 identifies an object corresponding to the overlap area. Probability values can be calculated. When the difference between the wavelength values of each object related to overlap is less than a certain value, the server 100 changes the wavelength value to show a difference of a certain amount or more by assigning a weight to at least one of the wavelength values corresponding to each object. You can do it.

In addition, the server 100 can identify moving objects by determining where each pixel point is located based on the calculated probability value corresponding to the overlap area and distinguishing the boundaries of each object based on the determination result. . For a specific example, if an overlap between wavelengths corresponding to two colors is identified and a probability value is calculated corresponding to the overlap area, it is determined whether each pixel point is in the direction of the first object or the direction of the second object based on the probability value. You can detect a moving object by checking in real time whether it exists. That is, the server 100 of the present invention can utilize the continuity of wavelengths to distinguish overlapping situations between objects and detect and track moving objects based on this.

Additionally, in an embodiment, the server 100 matches unique identification information related to a specific color to each object and performs object tracking based on the unique identification information, even when overlap between similar colors occurs. It can provide robust object detection and tracking capabilities.

In summary, the server 100 can distinguish a plurality of objects in an image based on the continuity of wavelengths corresponding to the plurality of colors present in the image, and when an overlap situation occurs between each object, a probability value related to the overlap area is calculated. The accuracy of object detection and tracking can be improved by distinguishing the boundaries between each object by calculating and using this to distinguish each pixel point. In addition, the server 100 determines the target continuity of each object by assigning unique identification information to each object, and based on the target continuity, predicts the movement of the target object even in overlapping situations between similar colors and reflects this to enable object tracking. Performance can be improved. The server 100 of the present invention can provide excellent detection and tracking performance regardless of changes in size, shape, occlusion, or rotation of the target object. A more detailed description of the method for detecting a moving object in an image based on wavelength information corresponding to color will be described later with reference to FIG. 3.

In the embodiment, only one server 100 in FIG. 1 is shown, but it is understood that more servers may also be included within the scope of the present invention and that server 100 may include additional components as set forth in the application. It will be clear to those skilled in the art. That is, the server 100 may be composed of a plurality of computing devices. In other words, a set of multiple nodes may constitute the server 100.

According to one embodiment of the present invention, the server 100 may be a server that provides cloud computing services. More specifically, the server 100 may be a type of Internet-based computing server that provides a cloud computing service that processes information not on the user's computer but on another computer connected to the Internet. The cloud computing service may be a service that stores data on the Internet and allows users to use it anytime, anywhere through Internet access without having to install necessary data or programs on their computer. The cloud computing service can be used to easily manipulate and manipulate data stored on the Internet. You can easily share and forward with a click. In addition, cloud computing services not only allow you to simply store data on a server on the Internet, but also allow you to perform desired tasks using the functions of applications provided on the web without having to install a separate program, and allow multiple people to view documents at the same time. It may be a service that allows you to work while sharing. Additionally, cloud computing services may be implemented in at least one of the following forms: Infrastructure as a Service (IaaS), Platform as a Service (PaaS), Software as a Service (SaaS), virtual machine-based cloud server, and container-based cloud server. . That is, the server 100 of the present invention may be implemented in at least one form among the cloud computing services described above. The specific description of the cloud computing service described above is merely an example, and may include any platform for constructing the cloud computing environment of the present invention.

The user terminal 200 according to an embodiment of the present invention may refer to any type of node(s) in a system that has a mechanism for communication with the server 100. The user terminal 200 may be a terminal that can receive detection and tracking information about a target object from image data acquired in real time through information exchange with the server 100. For example, the user terminal 200 may be a terminal related to a worker for security surveillance corresponding to a specific area. For another example, the user terminal 200 is related to a user who wants to obtain information related to the driving of a vehicle in the field of autonomous vehicles, or a user who wants to distinguish a specific target object from a medical image corresponding to a surgical procedure (e.g. , may be a terminal related to a specialist). As another example, the user terminal 200 may be a terminal related to a worker who evaluates and detects the appropriateness of motion of robot arms in the field of robot vision. The detailed description of the user terminal described above is only an example, and the present invention is not limited thereto.

In one embodiment, the user terminal 200 may be connected to the server 100 through the network 400, may provide real-time image data to the server 100, and may specify a moving object in response to the provided image data. You may be provided with identification information (or detection information).

User terminal 200 may refer to any type of entity(s) in the system that has a mechanism for communication with server 100. For example, the user terminal 200 includes a personal computer (PC), a note book, a mobile terminal, a smart phone, a tablet PC, and a wearable device. etc., and may include all types of terminals that can access wired/wireless networks. Additionally, the user terminal 200 may include an arbitrary server implemented by at least one of an agent, an application programming interface (API), and a plug-in. Additionally, the user terminal 200 may include an application source and/or a client application.

In one embodiment, the external server 300 may be connected to the server 100 through the network 400, and the server 100 performs a method of detecting a moving object in an image based on wavelength information corresponding to color. Various information/data required to do so can be provided, or the result data derived by performing a method of detecting a moving object in an image based on wavelength information corresponding to color can be provided, stored, and managed. For example, the external server 300 may be a storage server separately provided outside the server 100, but is not limited thereto.

In one embodiment, the external server 300 may be connected to the server 100 through the network 400, and the server 100 performs a method of detecting a moving object in an image based on wavelength information corresponding to color. Various information/data required to do so can be provided, or the result data derived by performing a method of detecting a moving object in an image based on wavelength information corresponding to color can be provided, stored, and managed. For example, the external server 300 may be a storage server separately provided outside the server 100, but is not limited thereto.

In various embodiments, information stored in the external server 300 can be used as learning data, verification data, and test data for training the neural network in the present invention. The external server 300 may be a server that stores a data set for training the deep learning model of the present invention. The external server 20 may be a digital device, such as a laptop computer, notebook computer, desktop computer, web pad, or mobile phone, equipped with a processor and equipped with memory and computing power. The external server 20 may be a web server that processes services. The types of servers described above are merely examples and the present disclosure is not limited thereto. Hereinafter, with reference to FIG. 2, the hardware configuration of the server 100 that performs a method of detecting a moving object in an image based on wavelength information corresponding to color will be described.

Figure 2 is a hardware configuration diagram of a server that performs a method of detecting a moving object in an image based on wavelength information corresponding to color related to an embodiment of the present invention.

Referring to FIG. 2, a server 100 that performs a method of detecting a moving object in an image based on wavelength information corresponding to a color related to an embodiment of the present invention includes one or more processors 110 and a processor 110. It may include a memory 120 that loads the computer program 151 executed by, a bus 130, a communication interface 140, and a storage 150 that stores the computer program 151. Here, only components related to the embodiment of the present invention are shown in Figure 2. Accordingly, anyone skilled in the art to which the present invention pertains will know that other general-purpose components may be included in addition to the components shown in FIG. 2.

According to one embodiment of the present invention, the processor 110 can typically process the overall operation of the server 100. The processor 110 provides or processes appropriate information or functions to the user or user terminal by processing signals, data, information, etc. input or output through the components discussed above or by running an application program stored in the memory 120. can do.

Additionally, the processor 110 may perform operations on at least one application or program for executing methods according to embodiments of the present invention, and the server 100 may include one or more processors.

According to one embodiment of the present invention, the processor 110 may be composed of one or more cores, such as a central processing unit (CPU) of a computing device, and a general purpose graphics processing unit (GPGPU). , may include a processor for data analysis and deep learning, such as a tensor processing unit (TPU).

The processor 110 may read a computer program stored in the memory 120 and provide a method of detecting a moving object in an image based on wavelength information corresponding to color according to an embodiment of the present invention.

In various embodiments, the processor 110 includes random access memory (RAM) (not shown) and read memory (ROM) that temporarily and/or permanently store signals (or data) processed within the processor 110. -Only Memory, not shown) may be further included. Additionally, the processor 110 may be implemented in the form of a system on chip (SoC) that includes at least one of a graphics processing unit, RAM, and ROM.

Memory 120 stores various data, commands and/or information. Memory 120 may load a computer program 151 from storage 150 to execute methods/operations according to various embodiments of the present invention. When the computer program 151 is loaded into the memory 120, the processor 110 can perform the method/operation by executing one or more instructions constituting the computer program 151. The memory 120 may be implemented as a volatile memory such as RAM, but the technical scope of the present disclosure is not limited thereto.

The bus 130 provides communication functions between components of the server 100. The bus 130 may be implemented as various types of buses, such as an address bus, a data bus, and a control bus.

The communication interface 140 supports wired and wireless Internet communication of the server 100. Additionally, the communication interface 140 may support various communication methods other than Internet communication. To this end, the communication interface 140 may be configured to include a communication module well known in the technical field of the present invention. In some embodiments, communication interface 140 may be omitted.

Storage 150 may store the computer program 151 non-temporarily. When performing a process to detect a moving object in an image based on wavelength information corresponding to color through the server 100, the storage 150 detects a moving object in the image based on wavelength information corresponding to color. Various information necessary to provide a process can be stored.

The storage 150 is a non-volatile memory such as Read Only Memory (ROM), Erasable Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), flash memory, a hard disk, a removable disk, or a device well known in the technical field to which the present invention pertains. It may be configured to include any known type of computer-readable recording medium.

The computer program 151, when loaded into the memory 120, may include one or more instructions that cause the processor 110 to perform methods/operations according to various embodiments of the present invention. That is, the processor 110 can perform the method/operation according to various embodiments of the present invention by executing the one or more instructions.

In one embodiment, the computer program 151 includes acquiring image data, generating a plurality of wavelength information corresponding to the image data based on a plurality of color information included in the image data, and generating a plurality of wavelength information corresponding to the image data based on the plurality of color information included in the image data. It may include one or more instructions for performing a method of detecting a moving object in an image based on wavelength information corresponding to color, including the step of detecting a moving object.

The steps of the method or algorithm described in connection with embodiments of the present invention may be implemented directly in hardware, implemented as a software module executed by hardware, or a combination thereof. The software module may be RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), Flash Memory, hard disk, removable disk, CD-ROM, or It may reside on any type of computer-readable recording medium well known in the art to which the present invention pertains.

The components of the present invention may be implemented as a program (or application) and stored in a medium in order to be executed in conjunction with a hardware computer. Components of the invention may be implemented as software programming or software elements, and similarly, embodiments may include various algorithms implemented as combinations of data structures, processes, routines or other programming constructs, such as C, C++, , may be implemented in a programming or scripting language such as Java, assembler, etc. Functional aspects may be implemented as algorithms running on one or more processors. Hereinafter, with reference to FIGS. 3 to 9, a method for detecting a moving object in an image based on wavelength information corresponding to color performed by the server 100 will be described in detail.

Figure 3 shows a flowchart illustrating a method for detecting a moving object in an image based on wavelength information corresponding to color related to an embodiment of the present invention. The order of the steps shown in FIG. 3 may be changed as needed, and at least one step may be omitted or added. That is, the following steps are only an example of the present invention, and the scope of the present invention is not limited thereto.

According to an embodiment of the present invention, a method of detecting a moving object in an image based on wavelength information corresponding to color may include acquiring image data (S100).

Image data of the present invention may be image data acquired in various fields, for example, in the field of autonomous driving, medical imaging, security, face recognition, and robot control. For example, the image data may include image data that is subject to automatic diagnosis, such as detection, segmentation, and quantification of lesions (eg, tumors or mutated tissues, etc.). Also, for example, the image data may include image data that is a target for object detection in an autonomous vehicle or drone. As another example, the video data may include video data that is subject to detecting signs of danger in the security industry. As another example, the image data may include image data that is used to help robots search for appropriate locations when assembling or repairing products in the manufacturing industry. The detailed description of the above-described image data is only an example, and the present invention is not limited thereto.

In an embodiment, acquiring image data may involve receiving or loading data stored in the memory 120. Acquisition of image data may include receiving or loading image data into another storage medium, another computing device, or a separate processing module within the same computing device based on wired/wireless communication means. For example, the server 100 may obtain image data from at least one of the user terminal 200 or the external server 300.

According to one embodiment, the server 100 may perform preprocessing on acquired image data. Here, preprocessing may include preprocessing related to morphology calculation to remove noise due to changes in illumination and movement.

Specifically, the server 100 may perform a morphology operation on image data to remove noise due to changes in illuminance and movement. The morphology operation may be an operation that removes noise by eroding and expanding the image through an operation using specific elements. By applying structural elements to the input image through morphological operations, an output image of the same size can be generated. In morphological operations, the value of each pixel in the output image is based on the result of comparing the pixel corresponding to the input image and neighboring pixels. The server 100 can remove noise caused by changes in illuminance and movement through expansion and erosion of morphology operations. Dilation can be adding pixels to the border of an object in an image, while erosion can be removing pixels from the border of an object. The number of pixels at which an image object is added or an image object is removed may vary depending on the size and shape of the structural elements used when processing the image. In morphological expansion and erosion operations, the state of each pixel in the output image is determined by applying a rule to the pixel corresponding to the input image and its neighboring pixels. The rules used when processing pixels can determine whether the association is dilation or erosion.

For example, in morphological operations, dilation may include a rule that sets the value of the output pixel to the maximum value among all neighboring pixels. For example, in the case of a binary image, if there is at least one neighboring pixel with a value of 1, the pixel is set to 1. Dilation of morphological operations can make an object visually more striking and fill in small holes in the object. In other words, lines can be expressed thicker and filled shapes can be displayed larger, making object distinction clearer.

For example, in morphological operations, erosion may include a rule where the value of the output pixel is the highest value among all neighboring pixels. For example, in the case of a binary image, if there is at least one neighboring pixel with a value of 0, the pixel is set to 0. Erosion of morphological operations removes floating pixels and thin lines, leaving only important objects. In other words, the remaining lines are expressed thinner and the shape is displayed smaller.

The server 100 of the present invention can remove small objects from images and smooth the edges of large objects by utilizing a combination of dilation and erosion of morphological operations, thereby minimizing noise due to changes in illumination and movement. You can.

According to an embodiment of the present invention, a method of detecting a moving object in an image based on wavelength information corresponding to color includes a plurality of wavelength information corresponding to the image data based on a plurality of color information included in the image data. It may include a generating step (S200).

According to embodiments, unique wavelength values may be different for various colors. For a specific example, referring to Figure 4, the wavelength corresponding to blue (B, blue) may be 454 nm, the wavelength corresponding to green (G, green) may be 555 nm, and red (R, red) The corresponding wavelength may be 660 nm. In addition to the corresponding color, various colors may have different wavelengths. For example, the wavelength corresponding to violet may be 410 nm, the wavelength corresponding to yellow may be 587 nm, and the wavelength corresponding to orange may be 600 nm.

As above, each color may have a different wavelength. The server 100 may generate a plurality of wavelength information corresponding to the image data based on a plurality of color information included in the image data through a configuration that generates different wavelength information for each color. That is, the server 100 can match wavelength information for each color corresponding to each pixel area of image data. Image data is an image acquired in real time and may include moving objects, and color changes may occur as time changes, and wavelength information may change in response to color changes. The server 100 can detect moving objects within image data based on changes in wavelength information.

According to an embodiment of the present invention, a method of detecting a moving object in an image based on wavelength information corresponding to color may include detecting a moving object based on a plurality of wavelength information (S300).

In one embodiment, a method of detecting a moving object in an image based on wavelength information corresponding to color may include identifying a plurality of objects present in image data based on a plurality of wavelength information.

According to one embodiment, the server 100 may perform segmentation on a plurality of objects based on wavelength information corresponding to a plurality of colors included in the image. Here, segmentation may mean classifying images into physically recognizable units of meaning. That is, the server 100 may perform segmentation on the video image based on the wavelength information of the video data. For example, as shown in FIG. 5, the server 100 stores the first object 201, the second object 202, the third object 203, and the fourth object 204 present in the image data. can be classified. As the server 100 distinguishes each object based on the wavelength of each color, it can not only distinguish the type of object but also match different labels to each object even if they are the same type of object. That is, as shown in FIG. 5, the four vehicles can be classified into individual objects.

Additionally, the server 100 may perform segmentation on each image constituting the image data and then identify the movement path of the classified object to generate movement path prediction information.

According to a specific embodiment, the step of detecting a moving object based on a plurality of wavelength information includes identifying a plurality of objects included in image data based on a plurality of wavelength information, each of the plurality of objects identified in the image data It may include the step of classifying and predicting the movement path of each of the plurality of divided objects based on changes in the plurality of wavelength information to generate movement path prediction information.

According to an embodiment, wavelength information is information that serves as a standard for distinguishing differences by color, so when segmentation is performed based on color wavelength information, object distinction can be made easier and classification accuracy can be improved. there is.

In other words, the server 100 identifies moving objects based on the wavelength information corresponding to the color of each object and classifies them into recognizable units (i.e., image-based segmentation), and each object classified through changes in wavelength Movement or movement can be detected to generate movement path prediction information corresponding to each object. The movement path prediction information for each object generated through the server 100 is used to predict the movement continuity of each object when overlap between each object occurs, thereby improving the classification accuracy of each object. In particular, when overlap occurs between objects of similar colors, detection and tracking can be performed by clearly distinguishing each object by continuously tracking the target object through the movement path prediction information of each object.

In one embodiment, a method of detecting a moving object in an image based on wavelength information corresponding to a color includes matching unique identification information corresponding to each of a plurality identified in the image data and wavelength information corresponding to each object. It may include tracking a moving object within the image data based on changes in and unique identification information.

According to one embodiment, when overlap occurs between a plurality of objects, the server 100 continuously tracks the target object through the movement path prediction information of each object, so that each object can be clearly distinguished and detected and tracked. there is. According to an embodiment, the server 100 may detect that overlap between a plurality of objects occurs based on the unique identification information of each object.

In a specific embodiment, the step of tracking a moving object within image data includes generating movement path prediction information for each object corresponding to unique identification information through a movement path prediction algorithm, and when overlap between a plurality of objects occurs. , may include tracking a moving object based on movement path prediction information corresponding to each object.

Here, the movement path prediction algorithm analyzes the movement pattern of the object based on analysis of previous frames of video data, and predicts the location where the object is likely to move in the next frame and the movement speed of the object based on the analyzed movement pattern. It may be characterized by generating movement path prediction information.

Through the movement path prediction algorithm, movement path prediction information can be generated for each object corresponding to each unique identification information. In this case, the movement path prediction information corresponding to the object is information about the movement direction and movement speed of the object. may include. For example, the path prediction information may include information that the object moves in the left direction and the moving speed is 30 km/h. The detailed description of the movement path prediction information described above is only an example, and the present invention is not limited thereto.

That is, when overlap occurs between a plurality of objects, the server 100 can track the objects through movement path prediction information corresponding to each object. This enables detection and tracking by clearly distinguishing each object, even when overlap between similar colors occurs.

According to various embodiments, when objects that were in an overlapping state are separated, the server 100 may update the movement path prediction algorithm based on the actual movement location and movement speed of each object.

In a specific embodiment, the step of performing an update to the movement path prediction algorithm includes, when a plurality of overlapping objects are separated from each other, generating object movement information based on the separation point of each object and the location of each object. , may include deriving a path prediction error by comparing object movement information and movement path prediction information corresponding to each object and updating a movement path prediction algorithm based on the path prediction error corresponding to each object.

In detail, when each object related to overlap is separated from each other, the server 100 may generate object movement information based on the separation point of each object and the location of each object. That is, the server 100 may generate object movement information based on the time when each object is separated and the location of each object at the time of separation. Object movement information includes information about the separation point between objects and information about the location of each object after separation, and may include information about the actual movement speed and movement position of each object.

In this case, the server 100 may derive a path prediction error by comparing the movement path prediction information predicted at the time when overlap occurs for a plurality of objects with the object movement information according to the movement result of the actual object. For example, the movement path prediction information predicted at the time when overlap between multiple objects occurs may include prediction information that the first object will move in the southeast direction at a speed of 20 km/h, and at the time when the overlap is canceled. Object movement information related to the corresponding actual movement result may include information that the object moved at a speed of 20 km/h in the northeast direction. The server 100 can identify that the amount of information is different and derive an error, and update the movement path prediction algorithm based on the derived error.

That is, the server 100 may perform an update to the movement path prediction algorithm by comparing the predicted movement path prediction information with the movement result of the actual object. Through this update process, the movement path prediction algorithm can more accurately predict the movement of the object in the next frame in response to the movement pattern of each object. In other words, if the prediction result is not accurate or there is an error in the prediction, the error can be derived by comparing it with the actual result, and the movement path prediction algorithm can be supplemented by receiving the error as feedback. As the movement path prediction algorithm improves and develops the shortcomings as described above, more accurate and efficient path prediction becomes possible, and can be further improved during continuous use.

In various embodiments, the server 100 may match unique identification information to each of a plurality of objects identified in the image data. Here, the unique identification information may be identification information randomly assigned to clearly distinguish each object. In embodiments, the unique identification information may include arbitrary wavelength information that matches differently to each of a plurality of objects. In one embodiment, arbitrary wavelength information may be generated corresponding to a wavelength value in a wavelength range that is not expressed as a color or a wavelength value corresponding to a rare color.

Additionally, in an embodiment, the server 100 may identify one object including various wavelength information and may match unique identification information corresponding to one identified object. For example, one object may include various colors rather than one color, and accordingly, a plurality of wavelength information may be generated corresponding to one object. In this case, since multiple wavelength information exists within one object, one object may be recognized as multiple objects, and an error may occur in object tracking. Accordingly, the server 100 can match one unique identification information to the object in order to recognize it by integrating it even if one object includes various wavelength information.

In a specific embodiment, the step of matching unique identification information includes identifying a plurality of wavelength information that are similar in direction and amount of change by a certain amount or more among adjacent wavelength information as the same object wavelength information, and uniquely identifying the same object wavelength information. It may include a step of matching information.

The server 100 can integrate wavelength information with similar change direction and amount of change between adjacent wavelength information and identify them as the same object wavelength information, even if the wavelength information is different from each other. Afterwards, the server 100 matches the unique identification information to the same object wavelength information integrated into the wavelength information corresponding to one object. Through this process, one unique identification information is matched to one object. The server 100 can detect and track moving objects based on unique identification information matched to each object. Since this is detection and tracking based on unique identification information given to each object, it can be characterized by having a solid recognition ability for objects even when overlap between similar colors occurs or when illuminance changes due to shadows or weather changes. . In particular, since target objects are detected and tracked based on unique identification information matched to each object, there is no need to perform a search corresponding to the entire area through a search window of a certain size. For example, when detecting an object using the accumulated value of the color histogram, a search window of a certain size must be applied from one location (e.g., the upper left end area) to the entire area, but the server 100 of the present invention Since objects are tracked based on unique identification information given to the object, continuous tracking is possible.

According to one embodiment of the present invention, when overlap between wavelength information occurs, the server 100 may detect a moving object using only the change in wavelength information corresponding to the overlap.

In a more specific embodiment, the step of detecting a moving object based on a plurality of wavelength information includes identifying at least two or more wavelength information related to overlap among the plurality of wavelength information, as shown in FIG. 6 ( S310) may be included. For example, as an object moves in the image data, overlap may occur between the first wavelength information 310 corresponding to the dotted line and the second wavelength information 320 corresponding to the solid line. This overlap occurs as the object corresponding to the first wavelength information 310 moves in the direction of the second wavelength information 320 (e.g., to the right), or the object corresponding to the second wavelength information 320 moves in the direction of the second wavelength information 320. This may occur as the 1-wavelength information 310 moves in the direction (eg, left direction).

Additionally, in an embodiment, the step of detecting a moving object based on a plurality of wavelength information may include generating corrected image data by removing objects that do not overlap from the image data (S320). Real-time overlap may occur when movement of an object occurs. Accordingly, the server 100 can generate corrected image data by removing parts not related to the moving object. This may be for detecting and tracking objects in motion.

Additionally, the step of detecting a moving object based on a plurality of wavelength information may include detecting a moving object based on wavelength information included in the corrected image data (S330). In other words, the server 100 can configure corrected image data by selecting only objects related to movement and detect moving objects based on the corrected image data.

According to one embodiment, the server 100 can identify the movement of each area through a change in wavelength and identify the occurrence of overlap between each area. Additionally, the server 100 may calculate a probability value corresponding to the intersection area of the two wavelength values and identify a moving object by distinguishing the location of each pixel point based on the calculated probability value.

In a specific embodiment, the step of detecting a moving object based on the wavelength information included in the corrected image data includes calculating a probability value based on the wavelength information included in the corrected image data and detecting the moving object based on the probability value. It may include a detection step.

Here, the probability value is a point that changes according to the movement of an object included in the image data, and may be characterized as being a standard for distinguishing the boundary of the wavelength. That is, the boundary between the two objects is distinguished depending on which direction each pixel point is located based on the probability value calculated corresponding to the overlapping area (eg, intersection area), and the movement of the object can be identified.

In an embodiment, the step of calculating a probability value includes identifying an intersection area 330 of first and second wavelength information related to overlap and calculating a probability value based on the center point of the intersection area 330. It can be included. Referring to FIG. 7 , the server 100 may calculate the center point of the intersection area 330 of the first wavelength information 310 and the second wavelength information 320 as a probability value 331. In other words, the probability value 331 may mean the center point of the intersection area 330 corresponding to overlap. This probability value 331 may continuously change according to the movement of the object, and the movement of the object may be detected by identifying the location of each pixel point based on the changing probability value 331.

In various embodiments, the step of detecting a moving object based on a probability value includes identifying which direction the location of each pixel is located in between the first wavelength information direction and the second wavelength information direction based on the probability value, each pixel It may include calculating the distance from the probability value in response to and detecting a moving object based on the identified location corresponding to each pixel.

Referring to FIG. 7, the server 100 classifies pixel points located in the direction of the first wavelength information 310 as the first object (or the first wavelength corresponding to the first object) based on the probability value 331. , Based on the probability value 331, pixel points located in the direction of the second wavelength information 320 can be classified as the second object (or the second wavelength corresponding to the second object).

Additionally, the server 100 may calculate the distance between each pixel point and the probability value based on the probability value 331. The distance from each pixel point to the probability value may be calculated as different values and may be a quantified factor. For example, arbitrary points a, b, and c may be located to the right based on the probability value 331, and points d, e, and f may be located to the left based on the probability value 331. In this case, the server 100 calculates the distance to each point based on the probability value 331 corresponding to each point. For a specific example, the distance between point a and the probability value 331 can be calculated as 10, the distance between point b and the probability value 331 can be calculated as 8, and the distance between point c and the probability value 331 can be calculated as 7.

As described above, moving objects are distinguished based on the location of each point based on the probability value 331 and the distance of each point calculated based on the probability value 331. That is, the server 100 detects a moving object using quantified factors such as the location of each point and the distance of each point corresponding to the probability value 331.

In other words, the server 100 of the present invention can track a moving object by distinguishing objects based on where each pixel point (i.e., pixel) is located based on the probability value 331. This is different from the technology that performs color-based object detection by utilizing the accumulation of normal distribution values of conventional pixel points, and determines the moving object by determining the direction in which each pixel point is located based on the overlapping area. can do.

In various embodiments, server 100 may generate a neural network model (eg, object detection model) that performs object detection. More specifically, referring to FIG. 8, the server 100 may obtain the pixel value for the focused area among the peak values (S410). Additionally, the server 100 may remove unnecessary areas (S420). That is, the server 100 can focus on an area where overlap occurs in the image data, obtain pixel values corresponding to the area, and remove unnecessary areas. By removing unnecessary areas, the ability to identify the area to focus on (i.e., the area where overlap occurs) is further improved, thereby improving the learning accuracy of the neural network.

Additionally, the server 100 may perform clustering on points in each direction based on the probability value 331 (S430). For a specific example, points a, b, and c located in the left direction (e.g., the first wavelength information direction) based on the probability value 331 may be clustered into the first cluster, and in the right direction based on the probability value 331. Points d, e, and f located in the second wavelength information direction can be clustered into a second cluster. That is, the server 100 can cluster each point into a plurality of clusters depending on which direction the points are located based on the probability value 331. In this case, the server 100 may generate learning data based on clustering (S440). In response to each cluster, one or more activation functions can be learned based on the learning data, and a neural network model for object detection can be created according to the learning results. The server 100 may calculate the object detection accuracy of the model after learning (S450). The server 100 may determine the performance of the generated model based on the detection accuracy of the model. According to the embodiment, the server 100 can create a plurality of artificial intelligence models by independently training each artificial intelligence model, and can evaluate performance and use only artificial intelligence models with a certain performance or higher.

According to an additional embodiment, when the difference between the wavelength information corresponding to the color of each object is less than a certain level (i.e., when the similarity between both wavelength information is more than a certain level), the server 100 weights at least one wavelength information to create an intersection area. It can be characterized by improving the recognition of .

More specifically, the step of calculating the probability value includes evaluating the similarity of the first wavelength information and the second wavelength information related to the overlap. If the similarity of the first wavelength information and the second wavelength information is greater than or equal to a predetermined similarity degree, the first wavelength information A step of assigning random wavelength information different from each other to each of the wavelength information and the second wavelength information, a step of identifying an intersection area between the random wavelength information assigned to each object, and a step of calculating a probability value based on the center point of the intersection area. It can be included.

For a specific example, referring to FIG. 9, since the colors of each of the first object (e.g., vehicle) 410 and the second object (e.g., building) 420 are similar to black, each object corresponds to The first wavelength information and the second wavelength information may also have a certain level or more of similarity. Figure 8 is image data acquired in real time from an autonomous vehicle, and may be image data used to detect and track various objects related to driving situations.

The first object 410 may sequentially move through the first and second positions to the third position. The first object 410 may overlap with a second object (eg, a building) 420 having a color similar to the object in the vicinity of the second location.

If the first object 410 completely overlaps the second object 420, it may become difficult for a vehicle driving while shooting and acquiring image data in real time to detect and track the first object. This may cause an accident due to the inability to detect objects ahead in an autonomous driving situation.

For example, when detecting an object based on the accumulated value of a histogram corresponding to a color, if the first object 410 completely overlaps the second object 420 because the pixel values of the corresponding area are the same, Clear detection and tracking of the first object 410 may be difficult. It is not simply limited to the overlap between similar colors between objects. If a dark-colored object passes through a shadow area that does not receive sunlight, the object may be obscured through the dark shadow area, making identification of the object difficult. This may make it difficult to identify the object. Errors may occur in object detection and tracking.

In an embodiment, when the color of each object is similar, the corresponding wavelength value is also similar, so an error may occur in the process of detecting the object based on the wavelength value. Accordingly, the server 100 may provide arbitrary wavelength information corresponding to each wavelength information in order to better distinguish the wavelength of each object. In one embodiment, arbitrary wavelength information may be generated corresponding to a wavelength value that does not exist in image data.

When overlap between objects is identified, the server 100 may identify whether the objects related to the overlap are objects corresponding to different wavelength information. Additionally, the server 100 may determine whether the difference in wavelength information of each object is greater than or equal to a predetermined standard value and assign a weight to at least one piece of wavelength information based on this. Specifically, if the difference in the wavelength information of each object is greater than a predetermined standard value, the server 100 may determine that the similarity of each wavelength information is below a certain level and calculate a probability value corresponding to the overlap area. In addition, when the difference in the wavelength information of each object is less than a predetermined standard value, the server 100 determines that the similarity of each wavelength information is above a certain level and assigns a weight (or random wavelength information) to each wavelength information to determine the similarity of each wavelength information. It can improve liver discrimination. In this case, the weights assigned to each wavelength information may be different.

For a specific example, the server 100 provides random wavelength information of 1000nm in response to the first wavelength information of the first object 410, and provides arbitrary wavelength information of 2000nm in response to the second wavelength information of the second object 420. Arbitrary wavelength information can be given. In one embodiment, the server 100 assigns arbitrary wavelength information (e.g., a wavelength value corresponding to a very high wavelength band) unrelated to the wavelength corresponding to a color as the wavelength information corresponding to each object to provide each wavelength information. It can improve the discernment between the liver. Through this, the server 100 can better identify the overlap area (or intersection area) of each wavelength information and calculate a probability value based on this. The server 100 can detect a moving object by identifying where each pixel point is based on a probability value calculated corresponding to the intersection area.

In other words, when overlap between objects with similar colors or wavelength information corresponding to colors occurs (i.e., overlap between wavelength information with a certain degree of similarity or higher), the server 100 assigns random wavelength information to each object to determine its wavelength information. By ensuring that the are clearly distinguished, the identification of the intersection area of each object can be improved. That is, the server 100 can weight the wavelength information corresponding to each object so that the intersection area can be identified more clearly. In addition, the improvement in the identification of the intersection area makes it possible to calculate accurate probability values, which makes it possible to detect and track moving objects by distinguishing the boundaries of each object. This has the advantage of providing robust object detection performance even when there is overlap between similar colors.

The steps of the method or algorithm described in connection with embodiments of the present invention may be implemented directly in hardware, implemented as a software module executed by hardware, or a combination thereof. The software module may be RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), Flash Memory, hard disk, removable disk, CD-ROM, or It may reside on any type of computer-readable recording medium well known in the art to which the present invention pertains.

The components of the present invention may be implemented as a program (or application) and stored in a medium in order to be executed in conjunction with a hardware computer. Components of the invention may be implemented as software programming or software elements, and similarly, embodiments may include various algorithms implemented as combinations of data structures, processes, routines or other programming constructs, such as C, C++, , may be implemented in a programming or scripting language such as Java, assembler, etc. Functional aspects may be implemented as algorithms running on one or more processors.

Those skilled in the art will understand that various illustrative logical blocks, modules, processors, means, circuits and algorithm steps described in connection with the embodiments disclosed herein can be used in electronic hardware, (for convenience) It will be understood that the implementation may be implemented by various forms of program or design code (referred to herein as “software”) or a combination of both. To clearly illustrate this interoperability of hardware and software, various illustrative components, blocks, modules, circuits and steps have been described above generally with respect to their functionality. Whether this functionality is implemented as hardware or software depends on the specific application and design constraints imposed on the overall system. A person skilled in the art may implement the described functionality in various ways for each specific application, but such implementation decisions should not be construed as departing from the scope of the present invention.

The various embodiments presented herein may be implemented as a method, apparatus, or article of manufacture using standard programming and/or engineering techniques. The term “article of manufacture” includes a computer program, carrier, or media accessible from any computer-readable device. For example, computer-readable media include magnetic storage devices (e.g., hard disks, floppy disks, magnetic strips, etc.), optical disks (e.g., CDs, DVDs, etc.), smart cards, and flash memory. Includes, but is not limited to, devices (e.g., EEPROM, cards, sticks, key drives, etc.). Additionally, various storage media presented herein include one or more devices and/or other machine-readable media for storing information. The term “machine-readable media” includes, but is not limited to, wireless channels and various other media capable of storing, retaining, and/or transmitting instruction(s) and/or data.

It is to be understood that the specific order or hierarchy of steps in the processes presented is an example of illustrative approaches. It is to be understood that the specific order or hierarchy of steps in processes may be rearranged within the scope of the present invention, based on design priorities. The appended method claims present elements of the various steps in a sample order but are not meant to be limited to the particular order or hierarchy presented.

The description of the presented embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be applied to other embodiments without departing from the scope of the invention. Thus, the present invention is not limited to the embodiments presented herein, but is to be construed in the broadest scope consistent with the principles and novel features presented herein.

Claims (10)

A method performed on one or more processors of a computing device, comprising:
Acquiring image data;
generating a plurality of wavelength information corresponding to the image data based on a plurality of color information included in the image data; and
Detecting a moving object based on the plurality of wavelength information;
Including,
A method of detecting moving objects in an image based on wavelength information corresponding to color.
According to paragraph 1,
The step of detecting a moving object based on the plurality of wavelength information,
Identifying a plurality of objects included in the image data based on the plurality of wavelength information;
distinguishing each of the plurality of objects identified in the image data; and
generating movement path prediction information by predicting a movement path of each of the plurality of divided objects based on changes in the plurality of wavelength information;
Including,
A method of detecting moving objects in an image based on wavelength information corresponding to color.
According to paragraph 2,
The step of detecting a moving object based on the plurality of wavelength information,
Identifying at least two wavelength information related to overlap among the plurality of wavelength information;
generating corrected image data by removing objects that do not overlap from the image data; and
Detecting a moving object based on wavelength information included in the corrected image data;
Including,
A method of detecting moving objects in an image based on wavelength information corresponding to color.
According to paragraph 3,
The step of detecting a moving object based on the wavelength information included in the corrected image data is,
calculating a probability value based on wavelength information included in the corrected image data; and
Detecting a moving object based on the probability value;
Includes,
The probability value is a point that changes according to the movement of an object included in the image data, and serves as a standard for distinguishing the boundary of the wavelength.
A method of detecting moving objects in an image based on wavelength information corresponding to color.
According to paragraph 4,
The step of calculating the probability value is,
Identifying an intersection area of first wavelength information and second wavelength information related to overlap; and
calculating the probability value based on the center point of the intersection area;
Including,
A method of detecting moving objects in an image based on wavelength information corresponding to color.
According to clause 5,
The step of detecting a moving object based on the probability value is,
Identifying which direction of the first wavelength information direction and the second wavelength information direction the position of each pixel is located based on a probability value;
calculating a distance from the probability value corresponding to each pixel; and
Detecting a moving object based on the location and distance identified for each pixel;
Including,
A method of detecting moving objects in an image based on wavelength information corresponding to color.
According to paragraph 4,
The step of calculating the probability value is,
Evaluating the similarity of first wavelength information and second wavelength information related to overlap;
When the similarity between the first wavelength information and the second wavelength information is greater than a predetermined similarity degree, assigning different random wavelength information to each of the first wavelength information and the second wavelength information;
Identifying an intersection area between random wavelength information assigned to each object; and
calculating the probability value based on the center point of the intersection area;
Including,
A method of detecting moving objects in an image based on wavelength information corresponding to color.
According to paragraph 1,
The above method is,
performing preprocessing corresponding to the image data; It further includes,
The preprocessing is,
Including preprocessing related to morphology calculations to remove noise due to changes in illumination and movement,
A method of detecting moving objects in an image based on wavelength information corresponding to color.
A memory that stores one or more instructions; and
A processor executing the one or more instructions stored in the memory,
The processor executes the one or more instructions,
A server that performs the method of claim 1.
A computer program combined with a computer as hardware and stored on a computer-readable recording medium so as to perform the method of claim 1.