Document image analysis has become an increasingly important domain today due to the desire to reduce the amount of paper documents and archives. Optical character recognition (OCR) systems and document structure analyzers are the essential tools to achieve this task. It often seems that the document to be recognized is not positioned correctly on the flatbed scanner, especially when the document is from a book or magazine. . This results in a distorted scanned image that poses a real problem for document analysis, understanding, character segmentation and recognition. Aligning the input image is therefore a crucial step in understanding the document. In this report, we propose an alignment method based on Hough transform and filtering algorithms. Furthermore, the study of the characteristics of the characters is necessary if a more faithful reconstitution of the document is expected. This involves studying the characters' color, average boldness and skeleton. Furthermore, most optical character recognition systems are sensitive to the quality and size of the characters provided. Skew angle correction and binarization steps damage characters, especially small ones. To maximize the efficiency of character recognition, characters are enlarged and dimmed using the pixel art scaling algorithm. This report consists of 2 sections. First, the different algorithms proposed to estimate the skew angle of a document and second, the study of character characteristics. In this section, we propose a simple method to estimate the skew angle of a document based on the combination of the Sobel edge detection filter, a filtering algorithm and the Hough shape transform. This method is designed for rapid detection of li...... half of the paper ......too many pixels are eliminated during the filtering step, which will therefore not leave enough information for the Hough transform. The tolerance of this filtering algorithm depends on the window size and the threshold value. The larger the difference between the window size and the threshold value, the more tolerant the algorithm is.1.2.2 Grayscale Image Filtering Algorithm This grayscale filtering algorithm is designed to do the same type of filtering as the previous algorithm, but on grayscale images. The greatest difficulty, compared to binary images, lies in the distinction between the objects and the background. In binary images we are only dealing with true or false values, while in grayscale images the values ​​range from 0 to 255. For the rest of this paragraph we assume that the text has a darker color than the background.