如何使用ITextSharp在PDF中获取嵌入图像的分辨率 [英] How can I get the resolution of an embeded image in a PDF using ITextSharp

问题描述

我已经构建了一种方法，试图查看给定pdf中所有嵌入图像的分辨率是否至少为300 PPI（打印值）。它的作用是遍历页面上的每个图像，并将其宽度和高度与artbox的宽度和高度进行比较。如果每页只有一个图像，它可以成功运行，但是当有多个图像时，artbox大小包括所有图像，抛弃数字。

I've built a method that tries to see if the resolution of all the embedded images in a given pdf is at least 300 PPI (print-worthy). What it does is cycle through each image on a page, and compare its width and height to the artbox width and height. It works successfully if there is only one image per page, but when there are multiple, the artbox size includes all of the images, throwing the numbers off.

我希望有人可能知道如何获取绘制图像的矩形大小，以便我可以正确比较，或者如果有更简单的方法获取图像对象的PPI（因为它将以矩形呈现，而不是以原始形式呈现）。

I was hoping that somebody might have some idea of how to get the rectangle size the image is drawn in so I can compare correctly, or if there is an easier way to get the PPI of an image object (as it would be rendered in its rectangle, not in raw form).

这是上述方法的代码

    private static bool AreImages300PPI(PdfDictionary pg)
    {
        var res = (PdfDictionary)PdfReader.GetPdfObject(pg.Get(PdfName.RESOURCES));
        var xobj = (PdfDictionary)PdfReader.GetPdfObject(res.Get(PdfName.XOBJECT));
        if (xobj == null) return true;
        foreach (PdfName name in xobj.Keys)
        {
            PdfObject obj = xobj.Get(name);
            if (!obj.IsIndirect()) continue;
            var tg = (PdfDictionary)PdfReader.GetPdfObject(obj);
            var type = (PdfName)PdfReader.GetPdfObject(tg.Get(PdfName.SUBTYPE));
            var width = float.Parse(tg.Get(PdfName.WIDTH).ToString());
            var height = float.Parse(tg.Get(PdfName.HEIGHT).ToString());
            var artbox = (PdfArray) pg.Get(PdfName.ARTBOX);
            var pdfRect = new PdfRectangle(float.Parse(artbox[0].ToString()), float.Parse(artbox[1].ToString()),
                float.Parse(artbox[2].ToString()), float.Parse(artbox[3].ToString()));

            if (PdfName.IMAGE.Equals(type) && (width < pdfRect.Width*300/72 || height < pdfRect.Height*300/72)
                || ((PdfName.FORM.Equals(type) || PdfName.GROUP.Equals(type)) && !AreImages300PPI(tg)))
            {
                return false;
            }
        }
        return true;
    }

供参考，以下是调用它的方法：

for reference, here is the method that calls it:

    internal static List<string> GetLowResWarnings(string MergedPDFPath)
    {
        var returnlist = new List<string>();
        using (PdfReader pdf = new PdfReader(MergedPDFPath))
        {
                for (int pageNumber = 1; pageNumber <= pdf.NumberOfPages; pageNumber++)
                {
                    var pg = pdf.GetPageN(pageNumber);
                    if (!AreImages300PPI(pg))
                        returnlist.Add(pageNumber.ToString());
                }
        }
        return returnlist;
    }

感谢您提供的任何帮助。

Thanks for any help you can provide.

推荐答案

我可以让你走完一条完全不同的道路吗？您正在查看生活在全局文件中的图像，但您没有看到它们在页面中的使用方式。

Can I put you down a totally different path? You're looking at images that live in the global file but you're not seeing how they're used in a page.

iTextSharp有一个名为 iTextSharp.text.pdf.parser.PdfReaderContentParser 可以走 PdfReader 并告诉你有关它的信息。您可以通过实现 iTextSharp.text.pdf.parser.IRenderListener 界面。对于遇到的每个图像，将使用 RenderImage 方法/tree/trunk/src/core/iTextSharp/text/pdf/parser/ImageRenderInfo.cs\"rel =nofollow> iTextSharp.text.pdf.parser.ImageRenderInfo 对象。从这个对象，您可以获得实际图像以及当前转换矩阵，它将告诉您图像如何放入文档。

iTextSharp has a class called iTextSharp.text.pdf.parser.PdfReaderContentParser that can walk a PdfReader and tell you things about it. You can subscribe to information by implementing the iTextSharp.text.pdf.parser.IRenderListener interface. For each image that it encounters the RenderImage method of your class will be called with an iTextSharp.text.pdf.parser.ImageRenderInfo object. From this object you can get both the actual image as well as the current transformation matrix which will tell you how the image is placed into the document.

使用此信息您可以创建一个这样的类：

Using this information you could create a class like this:

public class MyImageRenderListener : iTextSharp.text.pdf.parser.IRenderListener {
    //For each page keep a list of various image info
    public Dictionary<int, List<ImageScaleInfo>> Pages = new Dictionary<int, List<ImageScaleInfo>>();

    //Need to manually change the page when using this
    public int CurrentPage { get; set; }

    //Pass through the current page units
    public Single CurrentPageUnits { get; set; }

    //Not used, just interface contracts
    public void BeginTextBlock() { }
    public void EndTextBlock() { }
    public void RenderText(iTextSharp.text.pdf.parser.TextRenderInfo renderInfo) { }

    //Called for each image
    public void RenderImage(iTextSharp.text.pdf.parser.ImageRenderInfo renderInfo) {
        //Get the basic image info
        var img = renderInfo.GetImage().GetDrawingImage();
        var imgWidth = img.Width;
        var imgHeight = img.Height;
        img.Dispose();

        //Get the current transformation matrix
        var ctm = renderInfo.GetImageCTM();
        var ctmWidth = ctm[iTextSharp.text.pdf.parser.Matrix.I11];
        var ctmHeight = ctm[iTextSharp.text.pdf.parser.Matrix.I22];

        //Create new key for our page number if it doesn't exist already
        if (!this.Pages.ContainsKey(CurrentPage)) {
            this.Pages.Add(CurrentPage, new List<ImageScaleInfo>());
        }

        //Add our image info to this page
        this.Pages[CurrentPage].Add(new ImageScaleInfo(imgWidth, imgHeight, ctmWidth, ctmHeight, this.CurrentPageUnits));
    }
}

它使用此助手类来存储我们的信息：

It uses this helper class to store our information:

public class ImageScaleInfo {
    //The page's unit space, almost always 72
    public Single PageUnits { get; set; }

    //The image's actual dimensions
    public System.Drawing.SizeF ImgSize { get; set; }

    //How the image is placed into the page
    public System.Drawing.SizeF CtmSize { get; set; }

    //Automatically calculate how the image is scaled
    public Single ImgWidthScale { get { return ImgSize.Width / CtmSize.Width; } }
    public Single ImgHeightScale { get { return ImgSize.Height / CtmSize.Height; } }

    //Helper constructor
    public ImageScaleInfo(Single imgWidth, Single imgHeight, Single ctmWidth, Single ctmHeight, Single pageUnits) {
        this.ImgSize = new System.Drawing.SizeF(imgWidth, imgHeight);
        this.CtmSize = new System.Drawing.SizeF(ctmWidth, ctmHeight);
        this.PageUnits = pageUnits;
    }
}

使用它非常简单：

//Create an instance of our helper class
var imgList = new MyImageRenderListener();

//Parse the PDF and inspect each image
using (var reader = new PdfReader(testFile)) {
    var proc = new iTextSharp.text.pdf.parser.PdfReaderContentParser(reader);
    for (var i = 1; i <= reader.NumberOfPages; i++) {
        //Get the page object itself
        var p = reader.GetPageN(i);

        //Get the page units. Per spec, page units are expressed as multiples of 1/72 of an inch with a default of 72.
        var pageUnits = (p.Contains(PdfName.USERUNIT) ? p.GetAsNumber(PdfName.USERUNIT).FloatValue : 72);

        //Set the page number so we can find it later
        imgList.CurrentPage = i;
        imgList.CurrentPageUnits = pageUnits;

        //Process the page
        proc.ProcessContent(i, imgList);
    }
}

//Dump out some information
foreach (var p in imgList.Pages) {
    foreach (var i in p.Value) {
        Console.WriteLine(String.Format("Image PPI is {0}x{1}", i.ImgWidthScale * i.PageUnits, i.ImgHeightScale * i.PageUnits));
    }
}

编辑

从@ BrunoLowagie的评论下面我已经更新了上面的内容以删除魔术72，并且实际上尝试查询文档以查看是否已被覆盖。不太可能发生，但是一两年内有人会发现一些模糊的PDF并且抱怨这段代码不能比安慰那么安全。

From @BrunoLowagie's comments below I've updated the above to remove the "magic 72" and actually try querying the document to see if this has been overridden. Very unlikely to happen but someone in a year or two will find some obscure PDF and complain that this code doesn't work so better safe than sorry.

这篇关于如何使用ITextSharp在PDF中获取嵌入图像的分辨率的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！