Rendering flawed image detection method based on image recognition

Abstract

The invention provides a rendering flawed image detection method based on image recognition. The method comprises the following steps of inputting a rendering flawed image to be detected; calculating a differential matrix of the rendering image; determining the dimension of a detection template according to the rendering image change rate; calculating the neighbourhood minimum pixel difference of each pixel point in each frame of image according to the detection template; judging the located position of a broken frame image and flaws according to the minimum pixel differences. By aiming at the problems existing in the prior art, the invention aims at providing a video flaw detection method; flaw parts in the rendering result video image are fast detected.

Description

technical field

[0001] The present invention designs a rendering image defect detection technology, especially a method for rendering image defect detection. Background technique

[0002] Due to the large demand for computing resources and storage resources for home decoration and film and television rendering, the use of large-scale cloud computing technology to complete such rendering has become a mainstream technology. Therefore, the rendering system faces many different types of rendering tasks. During the resource loading process, there is a certain probability that some computing nodes cannot guarantee that all resources are loaded normally, which will cause some rendering nodes to render abnormally. When summarizing the rendering result image files, manual detection is used to identify the result images that have differences between the rendered scene and its adjacent images, define such images as abnormal frame images, and re-render them to correct errors. Although ...

Examples

Embodiment Construction

[0022] The present invention will be described in detail below in conjunction with various embodiments shown in the drawings.

[0023] refer to figure 1 Shown is a flow chart of a preferred system of the present invention, which is divided into input images and serialization processing, determination of single-pixel neighborhood templates, shifting and differential calculation of sequence image matrices, post-processing of differential modulus matrices, and statistical detection of bad Frame image and so on five steps.

[0024] refer to figure 2 The picture on the left is the picture of the normal rendering result, and the picture on the right is the picture of the abnormal rendering result. Because of the material loading problem, it can be clearly seen that the rendering of the grassland in the lower right corner of the right picture is abnormal, and there is a significant difference from the corresponding area in the left picture. The two frames of images are subtracted...

technical field

[0001] The present invention designs a rendering image defect detection technology, especially a method for rendering image defect detection. Background technique

[0002] Due to the large demand for computing resources and storage resources for home decoration and film and television rendering, the use of large-scale cloud computing technology to complete such rendering has become a mainstream technology. Therefore, the rendering system faces many different types of rendering tasks. During the resource loading process, there is a certain probability that some computing nodes cannot guarantee that all resources are loaded normally, which will cause some rendering nodes to render abnormally. When summarizing the rendering result image files, manual detection is used to identify the result images that have differences between the rendered scene and its adjacent images, define such images as abnormal frame images, and re-render them to correct errors. Although ...

Examples

Embodiment Construction

[0022] The present invention will be described in detail below in conjunction with various embodiments shown in the drawings.

[0023] refer to figure 1 Shown is a flow chart of a preferred system of the present invention, which is divided into input images and serialization processing, determination of single-pixel neighborhood templates, shifting and differential calculation of sequence image matrices, post-processing of differential modulus matrices, and statistical detection of bad Frame image and so on five steps.

[0024] refer to figure 2 The picture on the left is the picture of the normal rendering result, and the picture on the right is the picture of the abnormal rendering result. Because of the material loading problem, it can be clearly seen that the rendering of the grassland in the lower right corner of the right picture is abnormal, and there is a significant difference from the corresponding area in the left picture. The two frames of images are subtracted...

Images