Method for detecting workpiece based on homogeneous multi-core architecture and edge computing device

An edge computing device comprising:

• a processor;
• and
• a non-transitory storage medium coupled to the processor and configured to store a plurality of instructions, which cause the processor to:
• obtain detecting images of detecting workpieces;
• identify detecting areas of the detecting workpieces in the detecting images;
• perform a rotation operation on the detecting areas based on a preset rotation angle to obtain a rotation angle accuracy corresponding to the preset rotation angle and a target image similarity value, comprising:
• dividing the preset rotation angle to obtain the rotation angle accuracy and initial rotation angles;
• based on each of the initial rotation angles, rotating the detecting areas to obtain a rotation area of each of the initial rotation angles;
• calculating image similarity values between each of the rotation areas and a preset qualified area, and determining a largest image similarity value from the image similarity values as the target image similarity value, wherein an image similarity value represents image similarity between the detecting areas and a qualified area, and the rotation angle accuracy represents a ratio of the preset rotation angle to a preset value;
• and
• when the rotation angle accuracy is greater than or equal to a preset angle accuracy, identify whether the detecting workpiece is a qualified workpiece according to the target image similarity value and a preset image similarity threshold.

A non-transitory storage medium having stored thereon instructions that, when executed by at least one processor of an electronic device, causes the least one processor to execute instructions of a method for detecting workpiece based on homogeneous multi-core architecture, the method comprising:

• obtaining detecting images of detecting workpieces;
• identifying detecting areas of the detecting workpieces in the detecting images;
• performing a rotation operation on the detecting areas based on a preset rotation angle to obtain a rotation angle accuracy corresponding to the preset rotation angle and a target image similarity value, comprising:
• dividing the preset rotation angle to obtain the rotation angle accuracy and initial rotation angles;
• based on each of the initial rotation angles, rotating the detecting areas to obtain a rotation area of each of the initial rotation angles;
• calculating image similarity values between each of the rotation areas and a preset qualified area, and determining a largest image similarity value from the image similarity values as the target image similarity value, wherein an image similarity value represents image similarity between the detecting areas and a qualified area, and the rotation angle accuracy represents a ratio of the preset rotation angle to a preset value;
• and
• when the rotation angle accuracy is greater than or equal to a preset angle accuracy, identifying whether the detecting workpiece is a qualified workpiece according to the target image similarity value and a preset image similarity threshold.