Capsule head defect detection method and device

Abstract

The invention discloses a capsule head defect detection method and device, which belong to the field of image processing and pattern recognition. The method includes: acquiring a grayscale image of a capsule head to be detected; extracting a region of interest from the grayscale image, The region of interest contains bright spots; binarize the region of interest to obtain a binarized image; extract the edge of the image from the binarized image to obtain a target area of ​​bright spots; Extracting morphological features and / or grayscale statistical features from the target area; using a classifier to classify the morphological features and / or grayscale statistical features to detect whether there is a defect in the capsule head. Compared with the prior art, the present invention has the advantages of being able to effectively identify qualified capsule heads and defective capsule heads, and has higher accuracy.

Description

technical field [0001] The invention relates to the fields of image processing and pattern recognition, in particular to a capsule head defect detection method and device. Background technique [0002] In the production of medicines, powders or granules that are irritating to the esophagus and gastric mucosa, or medicines that taste bad and are easy to volatilize are often packaged in capsules. These medicines are packed in capsules, which not only protects the properties of the medicine from being damaged, but also Protects the digestive organs and respiratory tract. Removing the capsule shell may result in drug loss, drug waste, and reduced drug efficacy. In addition, some drugs need to be dissolved and absorbed in the intestines, and the capsule is a kind of protection to protect the drugs from being destroyed by gastric acid. Therefore, capsules are widely used in the production and processing of pharmaceuticals. [0003] In the actual production of capsules, there ar...

AI Technical Summary

Problems solved by technology

[0003] Capsules are characterized by many types and large quantities in actual production, and the finished capsules will inevitably have some defects, mainly including damage, over-length, over-short, surface spots, uneven thickness, etc.

At present, domestic pharmaceutical factories mostly rely on manual detection of defects in the production process of capsules. This method cannot meet the quality requirements of modern industrial production.

Although some pharmaceutical factories have introduced the image processing method into the capsule detection to solve the problems of manpower, material resources, secondary contact pollution, slow detection speed and inaccurate subjective judgment required by manual detection, they are all researched on the capsule body. , there are few research results on the defect detection of the capsule head, thus ignoring the damage at both ends of the capsule

Images