A Method for Inventorying Books on Shelves Based on Image Segmentation and Random Hough Transformation

Abstract

The invention discloses a method for counting books on a shelf based on image segmentation and random hough conversion, and belongs to the technical field of library management. First, according to edge pixel accumulation of an image, frame information is extracted, then the image is divided into multiple sub-images, and each sub-image represents a book spine image in a placement lattice on a shelf; then, through random hough mapping, rapid accumulation of book boundary straight lines in the sub-images is achieved in parameter space; finally, a local peak value is detected in the parameter space, in this way, the number of books in each placement lattice on the shelf is counted, and finally the number of all the books on the shelf is obtained. Advantages of image processing and advantages of random hough conversion are combined well; a computer is utilized for automatically counting the books on the shelf; the method is simple, efficient and easy to implement, greatly reduces the workload of book counting, and has high practicability and application and popularization value.

Description

technical field [0001] The invention belongs to the technical field of library management, and in particular relates to a method for counting books on shelves based on image segmentation and random hough transformation. Background technique [0002] Book counting is the basic data in book statistical indicators, and it is a difficult problem in library management. By counting books, the existing book property of the library can be accurately determined, which has important reference value for objectively reflecting the school's educational strength and striving for more funding. Therefore, book counting is very necessary for the daily management of the library, but the traditional manual counting method is time-consuming and laborious, and has disadvantages such as large labor demand and untimely information update. For library management, it is an important research direction to digitize and automate the traditional book counting work. In recent years, image processing tec...

AI Technical Summary

Problems solved by technology

[0008] The above method uses the method of image processing to realize the counting of single-row books and improve work efficiency; but it is mainly aimed at the image processing of simple single-row books, that is, only the books in a single book on the shelf can be counted , and now the number of books in the library is very large, and there are multiple book shelves on each shelf, the above method is difficult to efficiently count these massive books; on the other hand, the above method uses the method of edge projection detection to calculate the accumulation of text on the spine Peak value, this method has higher requirements on the background of the spine; when there are few characters on the spine of the book or the difference from the background is not large, it is easy to cause the book to be missed; when there are many characters and are distributed in two columns on the spine, it is easy to cause book misrepresentation

The occurrence of above-mentioned problem has limited the practicability and the range of application of background technology

Images