Data encryption method and apparatus, computer device, and storage medium
A data encryption method includes: acquiring target data to be encrypted of a target object and object information corresponding to the target object, the object information including at least one of a face image or an audio signal of the target object; encrypting an object feature corresponding to the object information according to a reference public key, and determining the encrypted object feature as a safety factor of the target data; and encrypting the target data based on the safety factor, to obtain encrypted data corresponding to the target data.
1. A data encryption method, performed by a computer device, the method comprising:
acquiring target data to be encrypted of a target object and object information corresponding to the target object, the object information comprising a face image and an audio signal of the target object;
obtaining an image feature of the face image and an audio feature of the audio signal of the target object respectively;
concatenating the image feature and the audio feature, and determining the concatenated feature as an object feature;
encrypting the object feature corresponding to the object information according to a reference public key, and determining the encrypted object feature as a safety factor of the target data; and
encrypting the target data based on the safety factor, to obtain encrypted data corresponding to the target data.
2. The method according to claim 1 , wherein the encrypting the target data based on the safety factor, to obtain encrypted data corresponding to the target data comprises:
encrypting the target data based on an encryption format corresponding to the target data and the safety factor, to obtain the encrypted data, wherein the encryption format indicates a relative position relationship between the safety factor and the target data in the encrypted data.
3. The method according to claim 1 , wherein the encrypting an object feature corresponding to the object information according to a reference public key, and determining the encrypted object feature as a safety factor of the target data comprises:
encrypting the object feature according to the reference public key when the object information belongs to a living category, and determining the encrypted object feature as the safety factor, wherein the living category represents that the object information is information about a living object.
4. The method according to claim 3 , wherein the object information comprises the face image, the method further comprises:
obtaining an image feature of the face image, and dividing the image feature into a plurality of first region features, wherein the face image comprises a plurality of first face regions, and each first region feature corresponds to a first face region in the face image;
determining a category to which the first face region corresponding to the each first region feature belongs based on the each first region feature;
determining a first ratio of first face regions belonging to the living category in the plurality of first face regions; and
determining that the face image belongs to the living category in response to the first ratio being greater than a first reference ratio.
5. The method according to claim 3 , wherein the object information comprises the audio signal, the method further comprises:
obtaining an audio feature of the audio signal, and dividing the audio feature into a plurality of first clip features, wherein the audio signal comprises a plurality of first audio clips, and each first clip feature corresponds to a first audio clip in the audio signal;
determining a category to which the first audio clip corresponding to the each first clip feature belongs based on the each first clip feature;
determining a second ratio of first audio clips belonging to the living category in the plurality of first audio clips; and
determining that the audio signal belongs to the living category in response to the second ratio being greater than a second reference ratio.
6. The method according to claim 5 , wherein the determining a category to which the first audio clip corresponding to the each first clip feature belongs based on the each first clip feature comprises:
determining, for the each first clip feature, the category to which the first audio clip corresponding to the each first clip feature belongs based on the first clip feature and the audio feature.
7. The method according to claim 3 , wherein the object information comprises the face image, the method further comprises:
classifying an image feature of the face image based on a first classification model, to obtain a category to which the face image belongs.
8. The method according to claim 3 , wherein the object information comprises the audio signal, the method further comprises:
classifying an audio feature of the audio signal based on a second classification model, to obtain a category to which the audio signal belongs.
9. The method according to claim 1 , wherein the object information comprises the face image, the object feature comprises an image feature of the face image, the method further comprises:
dividing the face image into a plurality of second face regions;
performing feature extraction on the plurality of second face regions respectively, to obtain a second region feature of each second face region; and
concatenating the plurality of obtained second region features, to obtain the image feature.
10. The method according to claim 9 , wherein the performing feature extraction on the plurality of second face regions respectively, to obtain a second region feature of each second face region comprises:
performing, for the each second face region, edge detection on the second face region to obtain an image gradient corresponding to the second face region, and determining the image gradient as the second region feature of the second face region, wherein the image gradient represents a change rate of any two adjacent pixel points in the second face region.
11. The method according to claim 1 , wherein the object information comprises the audio signal, the object feature comprises an audio feature of the audio signal, the method further comprises:
dividing the audio signal into a plurality of second audio clips;
performing feature extraction on the plurality of second audio clips respectively, to obtain a second clip feature of each second audio clip; and
concatenating the plurality of obtained second clip features, to obtain the audio feature.
12. A data encryption apparatus, comprising:
a processor and a memory, the memory storing at least one computer program, the at least one computer program being loaded and executed by the processor to implement:
acquiring target data to be encrypted of a target object and object information corresponding to the target object, the object information comprising a face image and an audio signal of the target object;
obtaining an image feature of the face image and an audio feature of the audio signal of the target object respectively;
concatenating the image feature and the audio feature, and determining the concatenated feature as an object feature;
encrypting the object feature corresponding to the object information according to a reference public key, and determining the encrypted object feature as a safety factor of the target data; and
encrypting the target data based on the safety factor, to obtain encrypted data corresponding to the target data.
13. The apparatus according to claim 12 , wherein the encrypting the target data based on the safety factor, to obtain encrypted data corresponding to the target data comprises:
encrypting the target data based on an encryption format corresponding to the target data and the safety factor, to obtain the encrypted data, wherein the encryption format indicates a relative position relationship between the safety factor and the target data in the encrypted data.
14. The apparatus according to claim 12 , wherein the encrypting an object feature corresponding to the object information according to a reference public key, and determining the encrypted object feature as a safety factor of the target data comprises:
encrypting the object feature according to the reference public key when the object information belongs to a living category, and determining the encrypted object feature as the safety factor, wherein the living category represents that the object information is information about a living object.
15. The apparatus according to claim 14 , wherein the object information comprises the face image, and the processor is further configured to perform:
obtaining an image feature of the face image, and dividing the image feature into a plurality of first region features, wherein the face image comprises a plurality of first face regions, and each first region feature corresponds to a first face region in the face image;
determining a category to which the first face region corresponding to the each first region feature belongs based on the each first region feature;
determining a first ratio of first face regions belonging to the living category in the plurality of first face regions; and
determining that the face image belongs to the living category in response to the first ratio being greater than a first reference ratio.
16. The apparatus according to claim 14 , wherein the object information comprises the audio signal, and the processor is further configured to perform:
obtaining an audio feature of the audio signal, and dividing the audio feature into a plurality of first clip features, wherein the audio signal comprises a plurality of first audio clips, and each first clip feature corresponds to a first audio clip in the audio signal;
determining a category to which the first audio clip corresponding to the each first clip feature belongs based on the each first clip feature;
determining a second ratio of first audio clips belonging to the living category in the plurality of first audio clips; and
determining that the audio signal belongs to the living category in response to the second ratio being greater than a second reference ratio.
17. The apparatus according to claim 16 , wherein the determining a category to which the first audio clip corresponding to the each first clip feature belongs based on the each first clip feature comprises:
determining, for the each first clip feature, the category to which the first audio clip corresponding to the each first clip feature belongs based on the first clip feature and the audio feature.
18. A non-transitory computer-readable storage medium, storing at least one computer program, the at least one computer program being loaded and executed by a processor to implement:
acquiring target data to be encrypted of a target object and object information corresponding to the target object, the object information comprising a face image and an audio signal of the target object;
obtaining an image feature of the face image and an audio feature of the audio signal of the target object respectively;
concatenating the image feature and the audio feature, and determining the concatenated feature as an object feature;
encrypting the object feature corresponding to the object information according to a reference public key, and determining the encrypted object feature as a safety factor of the target data; and
encrypting the target data based on the safety factor, to obtain encrypted data corresponding to the target data.