Image Processing System

Abstract

To provide an image processing system for an animal's perspective that can capture an image from the animal's perspective in a way that does not burden the animal and can deepen communication between humans and pets. [Solution] The image processing system 1 includes a wearable camera 20 that is attached to the lower leg of an owner walking an animal and captures images from the animal's point of view during the walk, and a user terminal 50 that receives the images captured by the wearable camera 20. The user terminal 50 includes an image processing unit 51 that processes the received images and displays them on the screen of the user terminal 50.

Description

[Technical Field]

[0001] The present invention relates to an image processing system. [Background technology]

[0002] Conventionally, a system that captures images from an animal's perspective and outputs the image data to a user terminal is known. For example, in the animal management system, program, and computer-readable storage medium of Patent Document 1, a device-side camera 31 of an animal-side device 10 attached to a collar 98 of an animal (dog) 99 captures images of the front during a walk and outputs the image data to a user terminal 50. The animal manager (dog owner) can view the images from the animal's perspective, deepening communication between humans and their pets. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2019-106908 Summary of the Invention [Problem to be solved by the invention]

[0004] However, because the animal management system in Patent Document 1 is designed to have a camera attached to the collar of the animal (dog), the weight of the camera is placed on the animal's (dog's) neck, which is a concern as it can be a significant burden, especially for small dogs.

[0005] An object of the present invention is to provide an image processing system from an animal's perspective that can capture images from the animal's perspective in a way that does not burden the animal, thereby deepening communication between humans and pets. [Means for solving the problem]

[0006] The image processing system for an animal's perspective according to the present invention includes a wearable camera attached to the lower leg of an owner walking an animal to capture images from the animal's perspective during the walk, and a user terminal that receives the images captured by the wearable camera. The user terminal includes an image processing unit that processes the received images and displays them on the screen of the user terminal.

[0007] Preferably, the image processing unit performs horizontal correction on the image captured by the wearable camera.

[0008] Preferably, the image processing unit performs image recognition of animals captured in images taken by the wearable camera, and marks and displays the animals. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram illustrating the configuration of an image processing system according to an embodiment. [Figure 2] FIG. 10 is a diagram showing an example of an image from a dog's perspective displayed on a user terminal. [Figure 3] (a) Image of image correction from downward to horizontal. (b) Image of image correction from upward to horizontal. [Figure 4] FIG. 10 is a diagram showing an example of an image in which another dog is detected and marked. DETAILED DESCRIPTION OF THE INVENTION

[0010] An image processing system according to one embodiment will be described with reference to the drawings. A typical pet is a dog (especially a small dog) that is kept indoors and needs to be taken for regular walks outdoors. Hereinafter, when distinguishing from a general "dog," the "owner's dog" will be referred to as the "pet dog."

[0011] 1, an image processing system 1 according to one embodiment includes a wearable camera 20 and a user terminal 50. Wearable camera 20 is attached to the lower leg, such as the ankle or shin, of an owner walking an animal, and captures images from the animal's perspective during the walk. The ankle or shin is merely an example, and the location of wearable camera 20 can be appropriately selected depending on the breed of dog, etc., so that it is positioned at a position close to the dog's eye level.

[0012] User terminal 50 receives the images captured by wearable camera 20. User terminal 50 is typically a PC or smartphone. User terminal 50 includes an image processing unit 51 that processes the received images and displays them on the screen of user terminal 50. Figure 2 shows an example of an image taken by wearable camera 20 and displayed on user terminal 50 from the perspective of a dog on a walk. Because the dog's line of sight is lower than that of a human, it is closer to the ground and roadside installations such as plants and utility poles appear to be more noticeable.

[0013] Next, horizontal image correction will be described with reference to FIG. 3. In the present invention, horizontal image correction refers to correcting an image so that it remains horizontal relative to the ground. For example, when going uphill or downhill, the image is corrected to be parallel to the gradient of the slope. Because wearable camera 20 is attached to the owner's lower leg, the angle of the lower leg changes as the owner walks, which may cause the image from wearable camera 20 to shake up and down. Therefore, image processing unit 51 of user terminal 50 horizontally corrects the image so that the height of the ground in the image remains constant, thereby suppressing image blur. For example, as shown in FIG. 3(a), if wearable camera 20 points downward rather than horizontally relative to the ground, image processing unit 51 corrects the image. Also, as shown in FIG. 3(b), if wearable camera 20 points upward rather than horizontally relative to the ground, image processing unit 51 corrects the image.

[0014] As shown in Fig. 4, the image processing unit 51 performs image recognition on an object, such as another dog, that appears in an image and marks it. In Fig. 4, the object is marked by displaying a frame around the four corners, but the marking may be displayed by enclosing it in a rectangle or circle, by using an arrow or triangle, or by other methods. Furthermore, the image processing unit 51 may learn to distinguish between objects in image recognition (for example, between a dog that is frequently encountered and other dogs, or between dogs and animals other than dogs (such as cats)), and mark them by distinguishing between their colors or types.

[0015] When the image is dark in a backlit scene or during a night walk, the image processing unit 51 may automatically correct the brightness when an object is recognized to make the object easier to see.

[0016] It is generally said that dogs can see fewer colors than humans can see. To more closely resemble the world a dog would see, the image processor 51 may display images captured in color in monochrome or in other tones (for example, tones without red).

[0017] After returning home from a walk, owners can view the recorded videos on their PC or smartphone. By reviewing the videos and reflecting on their dog's behavior during the walk, owners can understand their dog's feelings, the cause of any suspicious behavior, and notice changes in their dog's behavior that they may not have noticed during the walk. For example, it is difficult for a dog to see far from its point of view, and objects often appear relatively suddenly in front of them. Therefore, encountering a new dog (especially a large dog) can be quite a surprise for a dog. By understanding this by viewing the videos from the dog's point of view, owners can take into account the size of the dog and whether or not it is familiar to them, and choose behaviors that will minimize the shock of the encounter for their dog. Furthermore, if their dog tries to choose a different path than usual, owners can check what they saw from their dog's point of view in the videos and notice objects that caught their dog's attention that would not have been apparent from a human perspective.

[0018] By viewing images from the dog's perspective, owners can relive the emotional changes their dog may have experienced, allowing them to empathize with their dog's behavior and gain a deeper understanding of their dog's personality. For example, when choosing a walking route, owners can avoid the types of paths their dog dislikes and instead take them through areas with many things that interest their dog. This will help improve their dog's trust in their owner and build an even better relationship.

[0019] It is also possible to view the captured images with family members who are not out walking the dog. By sharing the dog's walk with other family members, it is possible to deepen communication not only between the dog and the family, but also between the dog walker and other family members. In addition, family members at home and other people other than the owner can view the images displayed in real time on their PCs or smartphones.

[0020] The conventional method of attaching a camera to a pet places a physical burden on the pet due to the weight of the camera, and may also place a mental burden on the pet due to having a foreign object attached to the body. In contrast, this embodiment uses a method in which the owner, not the pet, wears the camera, making it possible to capture images from the animal's point of view without placing any burden on the pet.

[0021] (Other embodiments) (a) The user terminal 50 may include an identification information registration unit, and may be able to register identification information (such as name and age) of an object that has undergone image recognition. When an object has a registered name, the name is displayed on the screen when the image is played back.

[0022] (b) The user terminal 50 may include a data processing unit, process the stored image data of the recognized objects, and display the images in order of the frequency of encounters.

[0023] (c) User terminal 50 may include a location information acquisition unit that acquires location information of wearable camera 20, and may combine image data of objects photographed by wearable camera 20 with the location information to display on a map where the object was encountered during a walk. User terminal 50 may also include a data processing unit that processes the accumulated image data, location information, and photographing time information, and may provide predictive information about where and what object is likely to be encountered on a route when walking along a route at a given time.

[0024] As described above, the present invention is not limited to the above-described embodiment, and can be implemented in various forms without departing from the spirit of the present invention. [Explanation of symbols]

[0025] 1. Image processing system, 20. Wearable Camera 50 User terminal, 51 Image processing unit

Claims

1. A wearable camera is attached to the lower leg of an owner walking an animal and captures images from the animal's perspective during the walk. a user terminal that receives images captured by the wearable camera; An animal-eye image processing system comprising: The user terminal is an image processing system including an image processing unit that processes the received image and displays it on a screen of the user terminal.

2. The image processing system according to claim 1 , wherein the image processing unit performs horizontal correction on the image captured by the wearable camera.

3. The image processing system according to claim 1 , wherein the image processing unit performs image recognition of animals captured in the image taken by the wearable camera, and marks and displays the animals.

Images