Picture detection method, device, equipment and medium

An electronic equipment, comprising:

• a memory, a processor, the processor being electrically connected to the memory, the memory storing a computer program executable by the at least one processor, the computer program, when executed by the at least one processor, causing the at least one processor to perform the steps of:
• acquiring a certificate picture uploaded by a user, and confirming a portrait direction on the certificate picture so as to obtain a correct state certificate picture;
• acquiring a highlight region of the correct state certificate picture;
• judging whether the highlight region is a light reflection region;
• if the highlight region is not a light reflection region, sending the correct state certificate picture to a server;
• if the highlight region is a light reflection region, acquiring a non-light reflection region picture of the correct state certificate picture, and identifying text region defect features and portrait region defect features in the non-light reflection region picture;
• using a pre-trained text region boundary identification model to identify a complete text region in the correct state certificate picture according to the text region defect features, and using a pre-trained portrait region boundary identification model to identify a complete portrait region in the correct state certificate picture according to the portrait region defect features;
• dividing a key information region of the correct state certificate picture according to the complete text region and the complete portrait region so as to obtain a key information picture, and sending the key information picture to a server.

A non-volatile computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by one or more processors, causes the one or more processors to perform the steps of:

• acquiring a certificate picture uploaded by a user, and confirming a portrait direction on the certificate picture so as to obtain a correct state certificate picture;
• acquiring a highlight region of the correct state certificate picture;
• judging whether the highlight region is a light reflection region;
• if the highlight region is not a light reflection region, sending the correct state certificate picture to a server;
• if the highlight region is a light reflection region, acquiring a non-light reflection region picture of the correct state certificate picture, and identifying text region defect features and portrait region defect features in the non-light reflection region picture;
• using a pre-trained text region boundary identification model to identify a complete text region in the correct state certificate picture according to the text region defect features, and using a pre-trained portrait region boundary identification model to identify a complete portrait region in the correct state certificate picture according to the portrait region defect features;
• dividing a key information region of the correct state certificate picture according to the complete text region and the complete portrait region so as to obtain a key information picture, and sending the key information picture to a server.