Pet display system and pet display method

Abstract

The invention provides a pet display system and a pet display method. The system includes a display panel and a computing circuit. The computing circuit receives an input image and determines whether it is in a pet mode. If not in the pet mode, the computing circuit transmits the input image to the display panel, and the input image is displayed by the display panel. If in the pet mode, the calculation circuit reduces the red component of the input image to obtain a transformed image and transmits the transformed image to the display panel, and the transformed image is displayed by the display panel.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to a display system and a display method suitable for pet vision. BACKGROUND

[0002] With the change of social structure and family form, pets play an increasingly important role in modern families and are gradually regarded as one of the family members. Many pet owners not only pay attention to the health and nutrition of pets, but also attach great importance to their emotional needs and life quality. Therefore, more and more products and services related to pet interaction have emerged, such as food, toys, wearable devices, and even audio and video content and interactive devices, all of which are developing towards the direction of improving the well-being of pets and the experience of human-pet interaction.

[0003] However, existing display devices (such as televisions, tablets, mobile phone screens, etc.) are designed in terms of color and brightness based on human visual characteristics, and do not take into account the differences in visual perception of common pets such as dogs and cats. According to research, the color vision system of animals such as dogs and cats is different from that of humans. The content presented by traditional display devices may not provide clear or attractive visual stimuli for pets, thereby reducing their participation and interest in interaction. SUMMARY

[0004] The present disclosure proposes a pet display system, which includes a display panel and a computing circuit. The computing circuit receives an input image and determines whether it is in a pet mode. If it is not in the pet mode, the computing circuit transmits the input image to the display panel, and the display panel displays the input image. If it is in the pet mode, the computing circuit reduces the red component of the input image to obtain a transformed image, and transmits the transformed image to the display panel, and the display panel displays the transformed image.

[0005] In an embodiment of the present disclosure, the computing circuit described above is used to calculate a plurality of values of the input image in the L a b color space, and reduce the value in the a channel.

[0006] In an embodiment of the present disclosure, the computing circuit described above is used to reduce the value in the b channel.

[0007] In an embodiment of the present disclosure, the computing circuit described above is used to multiply the values in the L a b color space by a transformation matrix, thereby reducing the value in the a channel and increasing the value in the b channel.

[0008] In an embodiment of the present disclosure, the input image comprises a plurality of pixels, each pixel comprising a corresponding value. The computing circuit divides the pixels into a plurality of groups, obtains a representative pixel of a first group, multiplies the value comprised in the representative pixel by a transformation matrix to obtain a plurality of transformed values, and replaces the pixels of the first group with the transformed values.

[0009] In an embodiment of the present disclosure, the pet display system further comprises an image extraction device configured to obtain a plurality of pet images. The computing circuit calculates a gaze time based on the pet images, and determines the transformation matrix based on the gaze time.

[0010] In an embodiment of the present disclosure, the computing circuit is configured to obtain a set of preset variables of the transformation matrix, and set a plurality of sets of perturbation variables. The computing circuit adds each set of perturbation variables to the preset variables to generate a plurality of sets of test variables, and applies each set of test variables to the input image to obtain a corresponding gaze time. The computing circuit adjusts the set of preset variables based on the gaze time.

[0011] In an embodiment of the present disclosure, the computing circuit is configured to set a reward positively correlated with the gaze time, and adjust the preset variables according to the following mathematical formula.

[0012]

[0013] wherein is the preset variable, is a learning rate, and N is the number of perturbation variables, is a standard deviation corresponding to the perturbation variable, is the i-th set of perturbation variables, is a reward corresponding to the i-th set of perturbation variables.

[0014] In an embodiment of the present disclosure, the computing circuit is configured to set the reward according to the following mathematical formula. is the gaze time corresponding to the i-th set of test variables, and T is a real number.

[0015]

[0016] In an embodiment of the present disclosure, the pet display system further comprises an image extraction device and an interactive device. The image extraction device is configured to obtain a pet image. The interactive device is electrically connected to the computing circuit. The computing circuit is configured to determine whether the pet image belongs to a preset category, and to activate the interactive device if the pet image belongs to the preset category.

[0017] In another aspect, an embodiment of the present disclosure provides a pet display method. The pet display method includes receiving an input image, determining whether in a pet mode, if not in the pet mode, transmitting the input image to a display panel, displaying the input image by the display panel, and if in the pet mode, reducing a red component of the input image to obtain a transformed image, transmitting the transformed image to the display panel, and displaying the transformed image by the display panel.

[0018] In order to make the above features and advantages of the present disclosure more clear and comprehensible, specific embodiments are described below in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 is a schematic diagram illustrating a pet display system according to an embodiment.

[0020] Figure 2 is a block diagram illustrating a pet display system according to an embodiment.

[0021] Figure 3 is a flowchart illustrating a pet display method according to an embodiment.

[0022] Figure 4 is an example diagram illustrating a transformed input image according to an embodiment.

[0023] SYMBOL DESCRIPTION

[0024] 110: display device

[0025] 120: image extraction device

[0026] 130: interaction device

[0027] 140: pet

[0028] 150: electronic device

[0029] 160: user

[0030] 200: pet display system

[0031] 210: computing circuit

[0032] 220: display panel

[0033] 301-304: steps

[0034] 410: input image

[0035] 420: transformed image DETAILED DESCRIPTION

[0036] The following detailed description is provided to enable any person skilled in the art to make and use the present application. Various modifications to the embodiments described herein will be readily apparent to those skilled in the art, and the generic principles defined herein can be applied to other embodiments without the use of the inventive faculty. Therefore, the present application is not intended to be limited to the embodiments described herein, but is to be accorded the full scope that comprises all embodiments where visible light is used to display images to a pet.

[0037] As used herein, the terms "first", "second", etc. do not necessarily mean a sequential or chronological order, but can be used to distinguish an element from another element having a same technical term.

[0038] Figure 1 is a schematic diagram illustrating a context of a pet display system according to an embodiment. Referring to Figure 1 , the pet display system includes a display device 110, an image extraction device 120, and an interactive device 130. The context is applicable to a pet 140, which is a dog in this example, but can be a cat, a bird, or another animal in other embodiments.

[0039] The display device 110 has multiple modes, one of which is called a pet mode. When in other modes (e.g., a general mode, a movie mode, a game mode, etc.), the display device 110 displays images mainly for human viewing, and thus the colors of the displayed images are in accordance with human vision. However, when in the pet mode, the display device 110 displays images mainly for the pet 140 viewing, and thus the colors of the displayed images are in accordance with the pet 140 vision. For example, dogs mainly perceive blue and yellow colors, but have a weaker ability to distinguish red colors. Therefore, the red component can be reduced in the pet mode, thereby enabling the pet 140 to distinguish the content displayed by the display device 110. The interactive device 130 and the electronic device 150 are described in the following paragraphs.

[0040] Figure 2 is a block diagram illustrating a pet display system according to an embodiment. Referring to Figure 2 , the pet display system 200 includes a computing circuit 210, a display panel 220, the image extraction device 120, and the interactive device 130. In this embodiment, the computing circuit 210 and the display panel 220 are integrated in a display device 110, but in other embodiments the computing circuit 210 can be another device (e.g., a TV box). In this embodiment, the image extraction device 120 and the display device 110 are different devices, but in other embodiments the image extraction device 120 can be integrated with the display device 110.

[0041] The computing circuit 210 may include a central processing unit, microprocessor, microcontroller, image processing chip, deep-learning processing unit (DPU), neural network processing unit (NPU), tensor processing unit (TPU), application-specific integrated circuit (ASIC), or programmable logic device (PLD). In this embodiment, the computing circuit 210 is a time controller in the display device 110, but in other embodiments it may be other circuits with computing capabilities. The display panel 220 may be a liquid crystal display panel, an organic light-emitting diode panel, or an electronic paper panel. The image extraction device 120 may include a charge-coupled device (CCD) sensor, a complementary metal-oxide-semiconductor (CMOS) sensor, or other suitable photosensitive elements. In some embodiments, the image extraction device 120 may also include dual cameras, a structured light sensing device, a laser, or any element capable of sensing scene depth. Image extraction device 120 can be electrically connected to computing circuit 210 via wired or wireless means. Interactive device 130 can be an automatic pet feeder, pet button, toy, or other device that allows interaction with a pet. Similarly, interactive device 130 can be electrically connected to computing circuit 210 via wired or wireless means. Electronic device 150 is communicatively connected to computing circuit 210. Electronic device 150 is, for example, a mobile phone, laptop computer, personal computer, or other device with computing capabilities.

[0042] Figure 3 This is a flowchart illustrating a pet display method according to one embodiment. Please refer to it. Figures 1 to 3 Steps 301 to 304 are executed by the computing circuit 210, and will not be described in detail below. In step 301, an input image is received. This input image is an image prepared for display on the display panel 220, and the present invention does not limit the content of the input image.

[0043] At step 302, it is determined whether the pet mode is on. In some embodiments, the user can select one of the modes through an on-screen display. If the user selects the pet mode, step 302 determines that the pet mode is on. If the user selects another mode, step 302 determines that the pet mode is off. In other words, the computing circuit 210 can determine whether the pet mode is on according to the mode selected by the user. In other embodiments, the user 160 can send an instruction to the computing circuit 210 through the electronic device 150 to specify the pet mode. In some embodiments, the image extraction device 120 extracts an environment image, and the computing circuit 210 can detect whether there is a pet and a human in the environment image. If there is only a pet, the pet mode is on. Otherwise, another mode is on.

[0044] If the pet mode is off, step 303 is performed to send the input image to the display panel 220, and the input image is displayed by the display panel 220. In other words, the input image is displayed according to a preset color setting (consistent with human visual perception).

[0045] If the pet mode is on, step 304 is performed to reduce the red component of the input image to obtain a transformed image, and the transformed image is sent to the display panel 220, and the transformed image is displayed by the display panel 220. Figure 4 FIG. 4 shows an example of the transformation of the input image according to an embodiment. Referring to FIG. 4, the input image 410 is consistent with human visual perception. After the transformation of step 304, the transformed image 420 is obtained, which is consistent with pet visual perception. Figure 4

[0046] The input image can belong to the RGB, YUV, XYZ, or other color spaces. In this case, the color space is transformed first, and then the red component is reduced. For example, the input image can be transformed to the L a b color space. The L a b color space includes an L channel, an a channel, and a b channel, each of which has a corresponding value. In such an embodiment, for each pixel of the input image, the value of the a channel can be reduced to obtain the transformed image. There are various ways to reduce the value of the a channel. For example, the original value can be reduced by a fixed value. If the original value of the a channel is greater than 0, the value can be multiplied by -1 or set to 0.

[0047] ​In some embodiments, a When the channel value decreases, b will also increase. The channel values ​​are used to maintain a constant pixel brightness. In some embodiments, a The amount reduced by the channel will be the same as b. The amount by which the channel value increases. In other embodiments, a The amount reduced by the channel can be greater than or less than b. The amount by which the channel value increases.

[0048] In some embodiments, when the input image belongs to the RGB color space, the value of the red channel can be directly reduced without color space transformation. Alternatively, the values ​​of the green and blue channels can be increased simultaneously to maintain constant brightness.

[0049] In some embodiments, L can be used to reduce the red component. a b Multiplying a value in the color space by a transformation matrix, this calculation is expressed as the following mathematical formula 1.

[0050] [Mathematical Expression 1]

[0051]

[0052] Where X is a vector containing a pixel in L a b Three values ​​in the color space. M is the transformation matrix, with a size of 3. 3. Y is the transformed vector, and the values ​​contained in this vector are called the transformed values. The transformed image can be obtained by performing the calculation of Equation 1 on each pixel. The transformation matrix can be designed such that the transformed a... The channel value will decrease, and b The channel value will increase.

[0053] Performing the mathematical equation (Equation 1) on every pixel would consume too many computational resources. Therefore, some implementations can employ speedup methods. For example, vector quantization can be used, where multiple representative vectors are pre-defined, and then each pixel in the input image is calculated to determine which representative vector it is closest to, thus classifying the pixel as belonging to the corresponding representative vector. The representative vectors have a length of 3, and are L... Channel, a Channel and b The channel value. This represents the transformed vector obtained by the calculation in Equation 1, and all pixels classified into the same category will use this transformed vector.

[0054] In some embodiments, all pixels in the input image can be clustered into a plurality of clusters. A non-supervised clustering algorithm (e.g. K-means algorithm) can be employed in this process, and the present disclosure does not limit the number of clusters. A representative pixel can then be obtained for each cluster, for example, the centroid of the cluster can be taken as the representative pixel. The values contained in the representative pixel can then be multiplied by the transformation matrix to obtain transformed values, and all pixels in the cluster can be replaced by the transformed values. When the number of clusters is small, more computation can be saved but more original pixel information can be lost; on the contrary, when the number of clusters is large, more pixel information can be preserved but less computation can be saved. In some embodiments, the number of clusters can be determined according to the hardware resources.

[0055] The values in the transformation matrix M can be pre-set. In some embodiments, the values in the transformation matrix M can also be dynamically changed during execution. For example, please refer to Figure 1 With Figure 2 A plurality of pet images can be obtained by the image extraction device 120. The calculation circuit 210 can calculate a gaze time according to the pet images. The calculation circuit 210 can first detect the pupil of the pet 140, determine the gaze position according to the position of the pupil, and then calculate the time for which the gaze position stays on the display device 110 as the gaze time.

[0056] Next, the transformation matrix can be determined according to the gaze time. Specifically, a set of pre-set variables of the transformation matrix is first obtained, denoted as Then, a plurality of sets of perturbation variables are set, denoted as , represents the i-th perturbation variable, and the perturbation variable is a 9-length vector. In some embodiments, the perturbation variables can be sampled according to a Gaussian distribution, and the mean value of the Gaussian distribution is 0 and the standard deviation is Next, the pre-set variables and each perturbation variable are added to obtain a plurality of sets of test variables, denoted as the following mathematical formula 2.

[0057] [mathematical formula 2]

[0058]

[0059] where N is a positive integer, representing the number of all perturbation variables. is the i-th set of test variables. represents the i-th set of perturbation variables. Each set of test variables can form a transformation matrix M. Next, each set of test variables are applied to the input image (e.g., the transformation of mathematical expression 1) to obtain a transformed image, and after the transformed image is displayed, the corresponding gaze time is calculated by the above-mentioned method of detecting the pupil. For example, each set of test variables are applied for 10 minutes, and then the corresponding gaze time for each set of test variables is calculated. If the gaze time is large, it indicates that the corresponding test variable is more in line with the current pet's vision.

[0060] The gaze time can be used to adjust the above-mentioned preset variables . In some embodiments, the calculation circuit 210 can set a reward that is positively correlated with the gaze time. For example, the reward can be set according to the following mathematical expression 3.

[0061] [mathematical expression 3]

[0062]

[0063] wherein is the gaze time corresponding to the ith set of test variables. T is a real number. is the reward corresponding to the ith set of test variables. In other words, when the gaze time is less than or equal to the real number T, the reward is 0. When the gaze time is greater than the real number T, the reward is greater than 0. Next, the above-mentioned preset variables can be adjusted according to the following mathematical expression 4.

[0064] [mathematical expression 4]

[0065]

[0066] wherein is the learning rate. is the standard deviation of the plurality of sets of perturbation variables. Thus, when a set of test variables causes a longer gaze time, the preset variable will be adjusted to be closer to this test variable. Through such a method, the transformation matrix M (i.e., the adjusted preset variable) can be dynamically determined when facing different pets, and the transformation matrix M can be determined according to the gaze time to conform to the vision of the pet.

[0067] On the other hand, after the pet image is obtained by the image extraction device 120, the mood, action, etc. of the pet 140 can also be recognized by the pet image. When the computing circuit 210 judges that the pet image belongs to a preset category, the interactive device 130 is activated. The preset category can represent hunger or boredom, etc. and the interactive device 130 can be a toy or an automatic feeder. For example, when the pet image is detected to belong to the category of hunger, the computing circuit 210 can actively activate the automatic feeder. Or when the pet image is detected to belong to the category of boredom, the computing circuit 210 can actively activate the toy.

[0068] In other embodiments, the pet display system 200 can also include a microphone to extract a sound signal. The computing circuit 210 can judge whether the pet is bored or hungry according to the sound signal and the pet image.

[0069] In some embodiments, the user 160 can receive the image extracted by the image extraction device 120 through the electronic device 150. The electronic device 150 can also extract images about the user 160, which can be transmitted to the display device 110. In other words, the user 160 can make a video call with the pet through the display device 110 and the electronic device 150.

[0070] In some embodiments, the user 160 can also remotely control the interactive device 130 through the electronic device 150, thereby interacting with the pet 140. For example, the interactive device is a feeder, and the user 160 can transmit a message to the computing circuit 210 through the electronic device 150, and the computing circuit 210 activates the interactive device 130.

[0071] In some embodiments, the interactive device 130 is a button, and the content displayed by the display panel 220 is about a game. The pet can watch the game screen and play the game by pressing the button. Since the content displayed by the display panel 220 in this embodiment conforms to the pet's vision, the pet can have a better game experience.

[0072] Although the present application has been disclosed in the above embodiments, it is not intended to limit the present application, and those skilled in the art can make some changes and modifications without departing from the spirit and scope of the present application, and the protection scope of the present application shall be subject to the appended claims.

Claims

1. A pet display system, comprising: Display panel; and A computing circuit is used to receive at least one input image and determine whether it is in pet mode. If not in pet mode, the computing circuit transmits at least one input image to the display panel, where the display panel displays the at least one input image. In the pet mode, the computing circuit reduces the red component of the at least one input image to obtain at least one transformed image, and transmits the at least one transformed image to the display panel for display.

2. The pet display system of claim 1, wherein the computing circuit is configured to calculate a plurality of values ​​of the at least one input image in the L*a*b* color space, and reduce the value of the at least one input image in the a* channel.

3. The pet display system of claim 2, wherein the calculation circuit is used to increase the value in the b* channel of the value.

4. The pet display system of claim 3, wherein the computing circuit is used to multiply the value in the L*a*b* color space by a transformation matrix, thereby reducing the value in the a* channel and increasing the value in the b* channel.

5. The pet display system of claim 4, wherein the at least one input image comprises a plurality of pixels, each of the pixels comprising a corresponding value. The computing circuit is used to group the pixels to obtain multiple groups, obtain the representative pixel of the first group of the group, multiply the value contained in the representative pixel by the transformation matrix to obtain multiple transformed values, and replace the pixel of the first group with the transformed value.

6. The pet display system as described in claim 5, further comprising: Image extraction device, used to acquire multiple pet images. The computing circuit is used to calculate the gaze time based on the pet image and determine the transformation matrix based on the gaze time.

7. The pet display system as claimed in claim 6, wherein... This calculation circuit is used to obtain a set of preset variables for the transformation matrix and to set multiple sets of perturbation variables. The calculation circuit is used to add each of the multiple sets of perturbation variables to the set of preset variables to generate multiple sets of test variables, and to apply each of the multiple sets of test variables to the at least one input image to obtain the corresponding gaze time. The calculation circuit is used to adjust the set of preset variables according to the gaze duration.

8. The pet display system of claim 7, wherein the calculation circuit is used to set the reward to be positively correlated with the gaze duration, and to adjust the set of preset variables according to the following mathematical formula. in Preset variables for this group, Let N be the learning rate and N be the number of the multiple sets of perturbation variables. To account for the standard deviations of these multiple sets of disturbance variables, Let i be the i-th group of disturbance variables among these multiple groups of disturbance variables. This is the reward corresponding to the i-th group of perturbation variables.

9. The pet display system of claim 8, wherein the calculation circuit is configured to set the reward according to the following mathematical formula, in Let T be the gaze duration corresponding to the i-th test variable, where T is a real number.

10. The pet display system of claim 1, further comprising: Image extraction device for obtaining images of pets; Interactive device, electrically connected to the computing circuit, The computing circuit is used to determine whether the pet image belongs to a preset category; if so, the interactive device is activated.

11. A pet display method, executed by a computing circuit, The pet display method includes: Receive at least one input image; Determine if the device is in pet mode; If not in pet mode, the at least one input image is transmitted to the display panel for display; and the at least one input image is displayed on the display panel. In the pet mode, the red component of the at least one input image is reduced to obtain at least one transformed image, and the at least one transformed image is transmitted to the display panel for display.

12. The pet display method as described in claim 11, further comprising: Calculate multiple values ​​of the at least one input image in the L*a*b* color space, and reduce the value of the a* channel among the values.

13. The pet display method as described in claim 12, further comprising: Increase the value in the b* channel.

14. The pet display method as described in claim 13, further comprising: The value in the L*a*b* color space is multiplied by a transformation matrix, thereby reducing the value in the a* channel and increasing the value in the b* channel.

15. The pet display method of claim 14, wherein the at least one input image comprises a plurality of pixels, each of the pixels comprising a corresponding value, the pet display method further comprising: The pixels are grouped to obtain multiple groups. A representative pixel of the first group is obtained. The value contained in the representative pixel is multiplied by the transformation matrix to obtain multiple transformed values. The pixels of the first group are replaced with the transformed values.

16. The pet display method as described in claim 15, further comprising: Multiple pet images were obtained using an image extraction device; as well as The gaze time is calculated based on the pet image, and the transformation matrix is ​​determined based on the gaze time.

17. The pet display method as described in claim 16, further comprising: Obtain a set of preset variables for the transformation matrix, and set multiple sets of perturbation variables; Add each of the multiple sets of perturbation variables to the set of preset variables to generate multiple sets of test variables, and apply each of the multiple sets of test variables to the at least one input image to obtain the corresponding gaze time; as well as The preset variables are adjusted according to the gaze duration.

18. The pet display method as described in claim 17, further comprising: Set the reward to be positively correlated with the gaze duration, and adjust the set of preset variables according to the following mathematical formula: in Preset variables for this group, Let N be the learning rate and N be the number of the multiple sets of perturbation variables. To account for the standard deviations of these multiple sets of disturbance variables, Let i be the i-th group of disturbance variables among these multiple groups of disturbance variables. This is the reward corresponding to the i-th group of perturbation variables.

19. The pet display method as described in claim 18, further comprising: The reward is set according to the following mathematical formula. in Let T be the gaze duration corresponding to the i-th test variable, where T is a real number.

20. The pet display method as described in claim 11, further comprising: Pet images are obtained using an image extraction device; as well as Determine if the pet image belongs to a preset category; if so, activate the interactive device.

Images