TWI381322B - Imaging system and method - Google Patents

Description

Imaging system and imaging method thereof

The present invention relates to an imaging technique, and more particularly to an imaging system applied to an image capturing device and an imaging method thereof.

At present, when a user performs an on-site interview or shooting on a special occasion using an image capturing device such as a video recorder, it is necessary to blur the face of the interviewee in the played image to protect the interviewee. Generally, the image after the shooting is finished is blurred by the face and then played.

However, since the face of the captured image can only be blurred by the face, the user cannot immediately blur the face of the interviewee during the instant play, and the purpose of protecting the interviewee cannot be achieved.

In view of the above, it is necessary to provide an imaging system and an imaging method capable of performing face blurring immediately in real time on-site playback.

An imaging system includes a storage module, an identification module, an image acquisition module, and a processing module. The storage module is configured to store a reference image. The identification module is configured to identify and store a face reference feature vector of at least one target face from the stored reference image. The image acquisition module is configured to acquire an image from the outside. The identification module is further configured to identify a feature vector of each face from the acquired image. The processing module is configured to compare the identified face feature vectors with the stored face reference feature vectors to determine whether they are the same. If they are the same, change the face with the same face reference feature vector. The pixel value of each pixel in the determined region in the acquired image to be compared with the stored face reference feature Faces with the same vector are hidden.

An imaging method comprising the steps of: storing a reference image; identifying and storing a face reference feature vector of at least one target face from the stored reference image; acquiring an image from the outside; and identifying each face in the acquired image a feature vector; comparing each of the recognized face feature vectors with the stored face reference feature vector to determine whether the same is the same; if the same, changing the face with the same face reference feature vector to the acquired image The pixel values of the pixels in the determined area are hidden by the same face as the stored face reference feature vector.

Compared with the prior art, the imaging system and the imaging method hide the face that needs to be hidden in the acquired image, so that the user can protect the interviewee when playing the image on the spot.

The embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

1 is an imaging system 100 for an image capture device, which includes a storage module 110, an identification module 120, an image acquisition module 130, and a processing module 140. In the present embodiment, the imaging system 100 is applied to a digital video recorder.

The storage module 110 is configured to store a reference image. In the embodiment, in order to improve the processing efficiency of the imaging system 100, the user acquires the reference image in a preview state of the digital video recorder. It can be understood that the reference image can also be taken before the interviewed interviewee or obtained from other external electronic devices.

The identification module 120 is configured to identify and store a face reference feature vector T1 of a target face from the stored reference images. In this embodiment, the recognition module 120 identifies a face reference feature vector of the target face (the interviewee's face) from the reference image. The face reference feature vector T1 includes the face remaining unchanged in any state such as stationary or rotating, such as the ratio of the distance between the inner corner of the left/right eye and the outer corner of the eye, and the distance between the center of the two pupils and the center of the nose. One of a ratio of the ratio, the distance between the two corners of the mouth, and the distance between the centers of the two nostrils, or a combination of the ratios thereof. In the present embodiment, T1=d1/d2, where d1 is the distance between the two corners of the target face, and d2 is the distance between the centers of the two nostrils of the target face.

The image acquisition module 130 is configured to acquire an image from the outside. The image acquired by the image acquisition module 130 from the outside is a still image. The image acquisition module 130 includes components for sensing external light to acquire image functions and related circuits, such as a charge-coupled device (CCD), a complementary metal oxide semiconductor (CMOS), and the like.

The identification module 120 is further configured to identify a feature vector T2 of each face from the acquired image. In this embodiment, the feature vector T2=d3/d4 acquired by the recognition module 120, where d3 is the distance between the two corners of the face of the image, and d4 is the distance between the centers of the two nostrils of the face in the image.

The processing module 140 includes a comparison unit 142 and a processing unit 144. The comparing unit 142 is configured to compare the identified face feature vectors T2 with the stored face feature vectors T1 to determine whether they are the same. In the present embodiment, the comparison unit 142 compares the feature vector T2 of each face of the image with the stored face reference feature vector T1. The processing unit 144 is configured to change a pixel value of the same face in the acquired image in the same face as the face reference feature vector T1 to be compared with the stored face reference feature vector T1 The same face is hidden, such as changing the pixel value of the face in the captured image to 0 or 1, which turns the face area into black or all white so that the face is invisible in the acquired image. . In this embodiment, the processing unit 144 changes the pixel value of each pixel in the acquired image to the face identifier feature vector T1 that is the same as the pixel value in the region determined in the acquired image to 0. The face area becomes completely black.

Referring to FIG. 2, an imaging method of an imaging system 100 according to an embodiment of the present invention includes the following steps: S110: storing a reference image; S120: identifying and storing a facial feature of a target face from the stored reference image. a vector; S130: acquiring an image from the outside; S140: identifying a feature vector of each face from the acquired image; S150: comparing each identified face feature vector with the stored face reference feature vector to determine Whether they are the same; S160: If they are the same, change the pixel value of each pixel in the determined region in the acquired image with the same face reference feature vector to hide the same face as the stored face reference feature vector .

In step S110, the user acquires a reference image in a preview state of the digital video recorder. It can be understood that the reference image can also be taken before the interviewed interviewee or obtained from other external electronic devices.

In step S120, the recognition module 120 identifies and stores a face reference feature vector T1 of a target face in the stored reference image. In the present embodiment, T1=d1/d2, where d1 is the distance between the two corners of the target face, and d2 is the distance between the centers of the two nostrils of the target face.

In step S130, the image acquisition module 130 is used to acquire an image from the outside.

In step S140, the recognition module 120 identifies the feature vector T2 of each face from the acquired image. In the present embodiment, T2=d3/d4, where d3 is the distance between the two corners of the face of the face in the image, and d4 is the distance between the centers of the two nostrils of the face in the image.

In step S150, the comparison unit 142 compares the identified face feature vectors T2 with the stored face reference feature vectors T1 to determine whether they are the same. If they are the same, the process proceeds to step S150. If they are not the same, the process ends. Process.

In step S160, the processing unit 144 uses the same face as the face reference feature vector to change the pixel value of each pixel in the determined region to 0 to make the face region become full. black.

The imaging system and the imaging method utilize a processing unit when taking a picture or recording The face to be protected in the captured image is hidden, so that the user can protect the interviewee when playing the image on the spot.

In summary, the present invention complies with the requirements of the invention patent and submits a patent application according to law. However, the above description is only the preferred embodiment of the present invention, and the scope of the present invention is not limited to the above-described embodiments, and equivalent modifications or variations made by those skilled in the art in light of the spirit of the present invention are It should be covered by the following patent application.

100‧‧‧ imaging system

110‧‧‧Memory Module

120‧‧‧identification module

130‧‧‧Image Acquisition Module

140‧‧‧Processing module

142‧‧‧Comparative unit

144‧‧‧Processing unit

1 is a functional block diagram of an imaging system according to an embodiment of the present invention; and FIG. 2 is a flow chart of an imaging method of the imaging system of FIG. 1.

100‧‧‧ imaging system

110‧‧‧Memory Module

120‧‧‧identification module

130‧‧‧Image Acquisition Module

140‧‧‧Processing module

142‧‧‧Comparative unit

144‧‧‧Processing unit

Claims (7)

An imaging system includes a storage module, an identification module, an image acquisition module, and a processing module, wherein the storage module is configured to store a reference image, and the identification module is used to store the Identifying and storing at least one reference feature vector T1 of the target face in the reference image, the reference feature vector T1 of the target face is: T1=d1/d2, where d1 is the distance between the two corners of the target face, D2 is the distance between the centers of the two nostrils of the target human face, and the image acquisition module is configured to acquire images from the outside, and the identification module is further configured to identify the feature vector T2 of each face from the acquired image. The feature vector T2 of each face in the image is: T2=d3/d4, where d3 is the distance between the two mouths of the face in the image, and d4 is the distance between the centers of the two nostrils of the face in the image. The processing module is configured to compare the identified face feature vectors T2 with the stored face reference feature vectors T1 to determine whether they are the same. If they are the same, change the same as the face reference feature vectors T1. The face is determined in the acquired image Pixel values of pixels in the art will face with the stored reference feature vector of the face to hide the same T1. The imaging system of claim 1, wherein the processing module comprises a comparison unit and a processing unit, wherein the comparison unit is configured to respectively store the identified facial feature vectors T2 and the stored The face reference feature vector T1 is compared to determine whether the same is the same. The processing unit is configured to change the pixel value of each pixel in the region determined in the acquired image by the same face as the face reference feature vector T1. The face that is the same as the stored face reference feature vector T1 is hidden. The imaging system of claim 2, wherein the processing The unit changes the pixel value of each pixel in the region determined by the face identical to the stored face reference feature vector T1 in the acquired image to zero. The imaging system of claim 2, wherein the processing unit compares pixel values of respective pixels in a region determined by a face identical to the stored face reference feature vector T1 in the acquired image. Becomes 1. An imaging method includes the steps of: storing a reference image; and identifying and storing a face reference feature vector T1 of at least one target face from the stored reference image, the target face reference feature vector T1 is: T1= D1/d2, where d1 is the distance between the two corners of the target face, and d2 is the distance between the centers of the two nostrils of the target face; acquiring images from the outside; identifying the feature vector of each face in the acquired image T2, the feature vector T2 for identifying each face in the image is: T2=d3/d4, where d3 is the distance between the two corners of the face in the image, which is the distance between the centers of the two nostrils of the face in the image. Comparing each of the identified face feature vectors T2 with the stored face reference feature vector T1 to determine whether they are the same; if the same, changing the same face as the face reference feature vector T1 to the acquired image The pixel values of the pixels in the determined area are hidden by the same face as the stored face reference feature vector T1. The imaging method according to claim 5, wherein the pixel value of each pixel in the region determined by the face identical to the stored face reference feature vector T1 in the acquired image becomes zero. The imaging method according to claim 5, wherein the face of the acquired image that is the same as the stored face reference feature vector T1 is determined The pixel value of each pixel in the predetermined area becomes 1.