琦琦儿 2008-8-1 11:08
PS合成教程：合成神秘的面部图案
出自：网络
在此教程聂ainley各界人士，通过我们的过程中作出一个引人注目的形象。 This tutorial focuses on the big picture steps taken to create this image.本教程侧重于大图片采取的步骤，以建立此形象。 You'll learn some incredible techniques in this tutorial, and get a view into how Nik creates beautiful photo-manipulation based images.你将学到一些令人难以置信的技术在这方面的补习，并取得的看法如何聂创造美丽的照片操纵基于图象。 Let's check it out.让我们瞧瞧吧。
Final Image Preview最后图片预览
Before we get started, let's take a look at the image we'll be creating.之前，我们开始使用，下面我们来看一看在形象，我们将创造。 Click the screenshot below to view the full-size image.点击下面的截图，以查看完整大小的图片。 This is a unique tutorial from Nik Ainley of Shinybinary .这是一个独特的补习从聂ainley的shinybinary 。 It doesn't have an accompanying PSD file in the PSDTUTS Plus membership area, as the focus of this tutorial is different.它不具有陪同PSD的文件中的psdtuts加会员专区 ，作为重点本教程是不同的。 Nik explains the tutorial's focus in the following introduction.聂解释，补习社的工作重点在以下简介。
[img]http://www.missyuan.net/upimg/allimg/080801/1140340.jpg[/img]
Introduction导言
Hello everyone, before we get down to actually starting any Photoshop work I thought I would say a few words.大家好，在我们坐下来开始，其实在任何Photoshop的工作，我以为我会说几句话。 Since this is a fairly long article, there are some points I need to make on what to expect from it, how it differs from the majority of tutorials, and my philosophy on tutorials themselves.由于这是一个相当长的文章，也有一些点，我要对什么期望从它，它如何不同于大多数教程，和我的哲学对自己的教程。
First, and most importantly, I should state that this is by no means a tutorial for beginners.首先，最重要的是，我要国家，这是决不是一个补习初学者。 I would say that at least an average level of Photoshop knowledge is needed, as well as a fairly good understanding of Photoshop terminology.我可以说，至少有一平均水平的Photoshop知识是需要，以及作为一个较好的了解Photoshop的术语。 I will explain how to do things, or give keyboard shortcuts for the less common commands, but won't be explaining how to run a Gaussian blur, or how to select a layer.我将解释怎样做事，或给键盘快捷键为不太常见的命令，但不会说明如何执行高斯模糊，或者如何选择一个层。 If you need to know how to do these things you are reading the wrong tutorial!如果您需要知道如何做这些事情你是读错了补习！
Basically, the idea behind this tutorial is to document the process I used while making this image from start to finish.基本上，背后的想法本教程是文件的过程中我用的同时，使这一形象，从开始到结束。 Now this probably means it doesn't qualify as a traditional tutorial at all, as I had very little idea of what the end result would be when I started.现在，这可能意味着它没有资格作为一种传统的补习社的一切，正如我已经很少的想法是什么最终结果将是当我开始。 Consequently, there are steps taken to produce effects or changes that might not actually be included in the final image, as I decided to remove them at a later stage.因此，有采取的步骤产生影响或改变，可能实际上没有被包括在最后的形象，正如我决定，以消除他们在稍后的阶段。 I might also seem to do things in a rather odd order.我可能似乎也做一些事，而不是在一个奇数阶。 This order is just how things occurred to me, rather than every step having been planned prior to getting started.这项命令是如何发生的事情对我来说，而不是每一步都经过事先计划的入门。
Realistically, it would also be almost impossible for me to record every single action I make while producing an image.从实际出发，它也将几乎是不可能的，我记录每一个单一的行动，而我要生产1形象。 Not just because it would result in about 5000 steps, but also because it would completely disrupt my working process trying to note everything down.不仅因为这将导致约5000步骤，而且也因为它会完全破坏我的工作过程，试图注意的一切。 I'm sure like me, you would find it very difficult to create a cohesive image if you had to keep stopping every couple of seconds to write something down!我敢肯定，像我一样，你会觉得很难创造一个有凝聚力的形象，如果你曾保持停车每两秒钟写的东西！ I have however included all of what I consider to be the important steps and techniques.我有，但包括了所有的我认为是重要的步骤和技巧。
Another way in which this article might differ from a more traditional 'recipe' style tutorial is the lack of precise instructions.另一种方式在这条可能有所不同，从一种更传统的'方'作风，补习是缺乏精确的指示。 You won't find much like 'now add a glow of this exact color, this size, this opacity' or 'now run a filter with these exact settings.'你也不会觉得很像现在添加一个辉光这个确切的颜色，这种大小的，这浑浊的'或'目前运行的一个过滤器与这些确切的设置。 There are a couple of reasons for this.有一对夫妇的这种情况的原因。 The first is that I think this approach is unsuitable for an image like this that utilizes many different techniques.第一是，我认为这种做法是不适合的形象像这样利用许多不同的技术。 It's fine for a method on how to reproduce an exact text style, or icon design, but not here.它的优良一个方法就如何重现一的确切文字的风格，或图示的设计，但不是在这里。
Secondly, I feel that many decisions such as what color to use, how light something should be, or how blurry is really down to individual artist's taste.其次，我觉得很多的决定，如什么颜色的使用，如何轻一些应，或如何模糊，实在是下降到个别艺术家的味道。 I could tell you the exact hex codes of every color I've used, but I'm sure that everyone would have different ideas of what they think looks good.我可以告诉你确切的十六进制码的每一个颜色我已使用，但我敢肯定，大家会有不同的想法是什么，他们认为看起来很不错。 I try to give instructions where I think they are needed for the process being discussed.我尝试给予指示的地方，我认为他们是需要的过程中正在讨论中。
So in summary, this isn't the sort of tutorial you can follow like a cookbook and end up with the same image as me.所以总括来说，这不是排序补习社的您也可以按照就像一个菜谱，并最终具有相同的形象，作为我。 However, I hope that by explaining my methods you will be able to pick up some Photoshop skills, and learn about parts of the program you didn't know before.不过，我希望解释我的方法，您将可以拾取一些Photoshop的技能，并了解部分的程序，你不知道前。
Since, during the course of producing this image I will be using similar methods repeatedly, I will only give a full explanation the first time.因为，过程中产生这种形象，我会用类似的方法反复，我只作出全面的解释，第一次。 Rather than repeat the steps I use for eg photo extraction ten times, I will describe them once, and from then on simply say 'extract the photo.'而不是重复的步骤，我使用例如为照片提取的10倍，我会形容他们一次，从此就简单地说， '提取物的照片。
Anyway, it's time to start thinking about creating this image.无论如何，它的时间，开始思考建立这个形象。 Here is the photo I started with; I got it from iStock .这里是照片我开始与我得到它从istock 。 I will be combining this with other photos which I will show as I introduce them.我会结合起来，这与其他的照片，我会显示，正如我介绍他们。
[img]http://www.missyuan.net/upimg/allimg/080801/1140341.jpg[/img]
With a photo-manipulation based image like this, finding good photos to work with is absolutely key.带照片的操作基于这样的形象，寻找良好的照片一起工作，是绝对的关键。 I can look at up to 10,000 photos just to find the right one that I am willing to pay for, which is of good quality.我可以看看高达10000照片只是为了找到合适的一，我愿意付出，这是品质好。 It doesn't matter how fancy your Photoshop skills are if you are working with poor photos.不要紧，如何花式您的Photoshop技巧如果您是工作与穷人的照片。 Spend the time here and it will show in your final image.花时间在这里，而且将显示在您的最后形象。 If you need somewhere to find photos, I find this site excellent for all sorts of graphic resources: Bluevertigo .如果您需要在某处找到的照片，我觉得这是很好的网站，为各种图形资源： bluevertigo 。
Right, enough of that.的权利，受够了这一点。 Let's go.让我们去。
Step 1第1步
Chances are you will need to do some work to your photo before starting, even if it is of high quality.机会是您将需要做一些工作，您的照片在开始之前，即使是高质量的。 My photo is of a good size to use as my canvas, so I am happy to just open it up in Photoshop and work on it directly.我的照片是一个良好的大小使用作为我的画布上，所以我很高兴向各位只需打开它在Photoshop和工作，它直接。 Its size is 2912 x 4368 pixels .其规模是2912 x 4368像素 。 Working at high resolution gives greater control, and will allow the image to be printed at a decent size in the future if needed.工作在高分辨率赋予更大的控制，以及将允许形象印在一个像样的规模，在未来如果需要。
The first thing that strikes me is the image is too dark.第一件事是我罢工的形象是太暗。 To sort this out I run a shadows/highlight adjustment on it to bring out the dark sides of the face without completely killing the contrast.排序这一点，我运行一阴影/高光调整，它带出黑暗的两岸面对没有完全杀死的反差。
Next I want to neutralize the color cast of the image slightly, so I perform a color balance adjustment .明年我想瓦解偏色的形象稍微，所以我执行色彩平衡的调整 。 I add some more blue, cyan and green into the image until I find the skin tone a more pleasing color.我想补充一些更蓝，青和绿色的形象进入，直到我觉得肤色更可喜的颜色。 Since I will be getting rid of the background, hair, and clothing, the face is really the only part that I am concerned with here.自从我将摆脱的背景，毛发和衣物，面对真的是只有一部分，我所关注的与在这里。
The final thing I do is to clone tool out his few freckles/spots.最后，我做的是克隆的工具，他的几个雀斑/点。 I find them distracting and unnecessary.我觉得他们的注意力和不必要的。 Use the healing tool if you like, but I find the clone tool a lot more reliable.使用愈合的工具，如果你喜欢，但我觉得克隆工具，有很多更可靠。
[img]http://www.missyuan.net/upimg/allimg/080801/1140342.jpg[/img]
Step 2第2步
Now for the almost inevitable step that every photo-manipulator hates, extraction.现在几乎不可避免的步骤，每一个照片机械臂讨厌，提取。 I want to cut out the background so it doesn't distract me.我想删掉的背景，因此它不会分散我。 I can replace it later if necessary.我可以取代它稍后如果有必要。 Now there are endless ways to extract an image, and which one to use really depends on what you are trying to cut out.现在有无穷的方法提取物的形象，和其中使用实际上取决于什么您正试图切断。 In this case, I have a fairly irregular shape but nothing overly complex.在这种情况下，我有一个公平的不规则形状，但没有过于复杂。 That means I can't just do it by hand.这意味着我不能只是做手。
First, I create a layer behind my photo and fill it with a bright color (yellow here) that contrasts sharply with it.首先，我创建一个层的背后我的照片，并填写与光明的颜色（黄色这里）形成鲜明对比，与它。 This is so that when I start cutting him out I will see very easily which pixels I am hiding.这是为了让当我开始切割出来，我会看到很容易，其中像素，我躲藏起来。 I then add a layer mask, zoom in, and start tracing around his face with a small brush.然后我添加了一层面具，在变焦，并开始追踪靠近他的脸与一小笔。 Since I plan to cover up the hair and neck, I just roughly cut these out for now.自从我计划来掩盖头发及颈部，我只是粗略这些削减为现在。 No point tracing carefully around something that's going to be hidden anyway.没有点跟踪小心靠近一点的要隐藏的，无论如何。
For a bit more detail on this method of extraction, here's a very old tutorial of mine that should still be helpful: Extraction为一点更详细的关于这个方法提取，这里有一个很旧的补习社的矿井，应该仍然是有帮助的： 萃取
[img]http://www.missyuan.net/upimg/allimg/080801/1140343.jpg[/img]
Step 3第3步
Once I had extracted what I wanted, I changed the yellow background layer to black as I wanted this image to have a dark background.有一次我已提取我想，我改变了黄色背景层，以黑色作为我想这个形象有一个黑暗的背景。 The face is leaning a bit so I wanted to rotate it to a better alignment.面对是一边倒一点，所以我要轮流到一个更好的调整。 To help I added a couple of guides.帮助我已将一对夫妇的指南。 Then I lined up the eyes and nose with them as best as I could.那么，我排队的眼睛和鼻子与他们最好的，正如我可以。
[img]http://www.missyuan.net/upimg/allimg/080801/1140344.jpg[/img]
Step 4第4步
I also wanted to turn the face to look at me so it is more symmetrical.我也希望把面子看我，这是更对称。 To do this I used a warp transform to pull the face into a better alignment.这样做，这是我用了一个经变换拉面对成为一个更好的调整。
[img]http://www.missyuan.net/upimg/allimg/080801/1140345.jpg[/img]
Step 5第5步
Now although I don't want the face to be perfectly symmetrical, I did find the right (his left) eye a bit too closed.现在，虽然我不想面对，以绝对对称的，我没有找到合适的（左）眼睛有点过于封闭。 I decided to replace it with a reflected version of the other eye.我决定要取代它反映了一版的其他眼。 To do this took a few steps.为此采取了几个步骤。
First, I duplicated the face and reflected it horizontally using the transform tool .首先，我重复面对和反映，它横向使用变换工具 。 I then dropped the opacity down and aligned the reflected eye over the one I wanted to replace before putting it back to 100% opacity .然后我下降的不透明和不结盟反射眼超过1我想，以取代之前，把它恢复至100 ％的不透明 。 I then added a 'reveal none' layer mask (hold down ALT while Clicking the add layer mask icon) .当时我说： '没有透露'层的面具（按住的同时，按Alt点击添加层的面具图标） 。 Then, using a big soft brush, I brushed white onto the mask over the old eye to reveal the new eye.然后，使用一个大的软刷，我刷上白面具超过旧的眼睛，揭示了新的眼睛。 This way I made sure the new eye blended in nicely.这样，我肯定新眼混纺在皆大欢喜。
PS I have added a gradient to the background here.的PS我增添了梯度的背景在这里。 There was no particular reason to this, just one of the many little things I do while making an image depending on how I feel.有没有特别的理由，只是其中的许多小东西，我手边的同时，作出了形象，视乎如何，我觉得。 You'll notice more small irrelevant changes like this throughout the article.您会发现更多的小不相干的变化，这样整个的文章。
[img]http://www.missyuan.net/upimg/allimg/080801/1140346.jpg[/img]
Step 6第6步
I also decided that his eyes were a little too puffy, and so smoothed the skin out just below them.我也决定，他的眼睛有点太puffy ，所以平滑的皮肤出仅低于他们。 To do this took a few steps as well.为此花了几个月的步骤，以及。
Basically, I cloned skin from his cheeks up to just below his eyes.基本上，我是克隆皮肤从他的脸颊上最多仅低于他的眼睛是雪亮的。 However, the skin on his cheeks is actually lighter, although it may not look it, so a straight clone would not work.不过，皮肤对他的脸颊，其实是较轻的，虽然它可能不看，所以一直克隆是行不通的。 The healing brush also failed here so it has to be done manually.愈合刷也没有在这里，所以它必须做的手工操作。 First, I duplicated the face layer, and very roughly cloned skin from his cheeks upwards so that it covered all the area I wanted to smooth.首先，我重复面对层，非常粗略克隆皮肤从他的脸颊向上，使其涵盖所有领域我想顺利。
Next I added a reveal none layer mask and dropped the opacity to 50% or so.接下来，我增加了一个显示无层口罩及不透明下降到50 ％左右 。 This opacity change is because I didn't want to completely remove the folds beneath the eyes, as that would look odd.这混浊的改变是因为我不想完全移除倍下方的眼睛，因为这将看多。 Next I brushed white onto the layer mask over the areas I wanted to affect.接下来，我刷上白层的面具超过领域我想影响。 This looked a little too light so I slightly darkened the top layer using a curves adjustment (pull the middle of the curve down slightly).这似乎有点太轻，所以我稍微黑暗的顶层使用曲线调整 （拉中曲线略有下降） 。 If there had been no difference in lightness between the two areas, I could have done this all in one step with the clone tool.如果没有差异，在亮度之间的两个领域，我可以做这一切在一，加强与克隆工具。 This method works where lightness/colour adjustments are needed though.这种方法工程，那里的亮度/彩色调整需要，虽然。
[img]http://www.missyuan.net/upimg/allimg/080801/1140347.jpg[/img]
Step 7第7步
Wow, so after all that the main image is basically ready for use (hah! I wish, as you will see I have to make more adjustments later, but it's good for the moment).哇，这么后，所有主要的形象基本上是准备使用（河！我想，正如你会看到我已作出更多的调整之后，但它的好，为的时刻） 。 Now it's time to bring in our next photo.现在的时间，使我们的下一张照片。 This is another high resolution image, approximately the same size as the first, from iStock .这是另一种高分辨率图像，大约有大小相同，作为第一，从istock 。
[img]http://www.missyuan.net/upimg/allimg/080801/1140348.jpg[/img]
Step 8第8步
As before, I needed to extract this photo to make it ready for use.一如以往，我需要提取此照片，使之随时准备使用。 With an image set against a flat background like this that job is slightly easier.与一的形象，设置对一个单位的背景，从这样的工作是相对容易一些。 I find the best way to do these extractions is to use the RGB channels .我觉得最好的方法做这些提取物是使用的RGB渠道 。 I took a look at these channels to see which had the most contrast between the background and object, in this case the blue channel.我一看这些渠道，看到了最之间的反差的背景和对象，在这种情况下，蓝色通道。
Next I duplicated this channel.接下来，我重复这个频道。 Then in the curves adjustment window , selected the set black point dropper, and chose a very light grey.然后，在曲线调整窗口 ，选定的一套黑点滴管，并选择一个很浅灰色。 This changed every pixel from that light grey and lighter to black.这改变了每一个像素由浅灰色和黑色打火机。 I tend not to choose the lightest grey possible, as this results in very jagged edges because you basically cause aliasing.我倾向于不选择最轻的灰色可能，因为这样的结果非常锯齿状边缘，因为你基本上事业走样。 To make sure that you are getting all of your image requires a little more work.为了确保您得到您所有的形象，需要多一点的工作。
Tip: If you do end up with jagged edges in your channel, there is a way to help smooth them out. 提示：如果您这样做，最终与锯齿状边缘，在您的频道，有一种方法，以帮助他们顺利。 Run a series of very light Gaussian blurs (1 pixel or less) on your channel. 运行了一系列非常轻的高斯模糊 （ 1像素或更少）在您的频道上。 Then using a levels adjustment , pull the two outer sliders towards the middle until your channel sharpens up again, but stop before it gets jagged. 然后使用水平的调整 ，拉两个外滑块对中，直到您的通道锐化起来，但停止之前，它得到锯齿状。 Be careful though as this method can cause the smoothing of corners as well so fine details might get cut out of your selection. 要小心，虽然这种方法可以导致平滑的弯道，以及使罚款的细节可能会切断对您的选择。
[img]http://www.missyuan.net/upimg/allimg/080801/1140349.jpg[/img]
Step 9第9步
Using the magic wand tool, I selected the big white areas outside and in the middle of the frame.使用魔术棒工具，我选择了大白色以外的地区，并在中东的框架。 I then inverted this selection, contracted by a couple of pixels, and filled the resulting selection with black.当时我倒这个选择，承包，由一对夫妇像素，并填补由此产生的选择与黑色。 I now had a perfect black shape in the parts of the image I want and white where I didn't want it.我现在有一个完美的黑色形状，在部分的形象，我想和白色的地方我不想它。 This is actually the exact opposite of what I want, as white represents the bits of the image that will be selected, so I had to invert this channel.其实，这是完全相反的是什么，我想，作为白色代表位的形象，将选定的，所以我不得不反转此频道。 I then CNTRL + Clicked on this channel to produce a selection.当时我cntrl +点击了这个频道，以产生一个选择。 Then making sure the RGB channel was active, copied and pasted into my main document.然后作出肯定的RGB通道活跃，复制并粘贴到我的主要文件。
[img]http://www.missyuan.net/upimg/allimg/080801/11403410.jpg[/img]
Step 10第10步
Ok, so I now had my second photo cut out and in my main document.确定，所以我现在有我的第二个照片剪下，并以我的主要文件。 The first job here was to clean up the extraction.第一份工作在这里被清理的提取。
Extraction can often leave a 'haloing' effect where the background is just visible around the edges of the object.提取常常可以留下一个' haloing '效果的地方背景，只是有形靠近边缘的对象。 You can try using a defringe command to clean this up, although I find the results of this only good on small objects.您可以尝试使用defringe命令，以清洁跟进，虽然我找到的结果，这只是良好的细小的物体。 On a very high resolution image like this, the best result is simply to cut off the outer 1/2 pixels .一个非常高分辨率图像一样，这一点，最好的结果只不过是为了切断外1 / 2像素 。 To do this I got a selection based on the shape of my object (CNTRL + Click on the layer thumbnail) , contracted by 2 pixels, and then added a layer mask.为此我得到遴选的基础上，形成我的对象（ cntrl +点击该层缩略图） ，收缩2像素，然后补充了一层面具。
I then brushed black onto this mask to remove little bits of the background that had been missed by the extraction.当时我擦肩而过黑到这个面具，以消除小钻头的背景下，已错过了由提取。 As shown here, a little bit of the white background can be seen through the object itself.作为在这里显示，一点点的白色背景可以看出，通过对象本身。 It's best to clean things like this at the start before you go duplicating the object a lot and have to fix them all later.它的最佳清洁的东西，像这样在开始之前，请重复对象很多，并已修复的所有稍后。
[img]http://www.missyuan.net/upimg/allimg/080801/11403411.jpg[/img]
Step 11第11步
Now I needed to decide where exactly I wanted to blend this image into the face.现在我需要决定在哪里，正是我想融入这个形象到面对的问题。 The first thing I did was to duplicate my new object (let's call it the frame from now on) and hid one of the layers.第一件事，我本来是要重复我的新对象（让的号召，它的框架内从现在起）和HID之一层。 It's good to keep a backup of important objects like this in case you need to use it again.它的好，以保持一个备份的重要对象，像这样的情况下，您需要再次使用它。
Next I chose a part of the frame I was interested in, and very roughly masked it off.接下来，我选择了部分的框架，我有兴趣，也很粗略蒙面将它关闭。 I then moved it around, resized and rotated it to see what it looked like in different positions.当时我提出它靠近，大小和旋转它，看看它看上去像在不同的位置。 I decided that one on either side of the face above the eyes looked good.我决定一的两旁，面对上述的眼睛看的好。
[img]http://www.missyuan.net/upimg/allimg/080801/11403412.jpg[/img]
Step 12第12步
Once I had decided on the part of the frame I wanted, I needed to cut it out properly.有一次我已决定对部分框架我想，我需要削减它妥善。 I got rid of the rough mask, added another, and this time traced around it carefully.我得到摆脱粗糙的面具，增加了一个，这时间追溯到周围仔细。
[img]http://www.missyuan.net/upimg/allimg/080801/11403413.jpg[/img]
Step 13第13步
Now here's a little trick that can be used when masking like this to get a clean result.现在这里有一个小动作，可以用来当掩蔽这样生成一个干净的结果。 Sometimes it can be difficult to tell whether you've left any stray pixels when masking, even if you look directly at the mask itself.有时它可以很难说是否有您留下的任何杂散像素时，掩蔽，即使你看看直接在面具本身。 By simply adding a stroke layer style (standard settings are perfect) these rogue pixels instantly shown up.由简单地增加中风层的建筑风格 （标准设置是完美的）这些流氓像素即时显示。 Even a single pixel which is only 1% opaque will have a stroke added making it obvious.即使是单一像素这是只有1 ％的不透明将有中风的补充，使显而易见的。 You can then quickly whizz over them with black on the layer mask.然后，您就可以迅速whizz超过他们对黑层的面具。
[img]http://www.missyuan.net/upimg/allimg/080801/11403414.jpg[/img]
Step 14步骤14
Ok, so I've got a nicely cut out object now, although the bottom is a little bit odd.确定，所以我提供了很好的削减了对象，现在，虽然底部是一个有点怪。 This is actually down to the image though as there was no nice edge to cut to.其实，这是下降到形象，虽然由于没有很好的优势，削减。
[img]http://www.missyuan.net/upimg/allimg/080801/11403415.jpg[/img]
Step 15第15步
To sort this bottom edge out I added a layer mask (after applying the previous one used to cut the object out), and started softly brushing black over the bottom area.排序这底边离开系统I增添了一层面具（后申请前1用来削减的对象列） ，并开始轻轻刷黑超过底部区。 I used the smudge tool to push the black upwards into the layer to make it slightly less uniform.我用了黑灰的工具，把黑色调高到层，使其略少制服。
This image also needed a bit of adjustment to match the overall lighting and colors of the image.此图片也需要一点调整，以配合整体照明和色彩的形象。 The steps I took to fix it are as follows.步骤，我曾来修复它是如下。 First, a very light shadows/highlights adjustment to bring out some of the darker areas.首先，一个很轻的阴影/突出了调整，带出了一些黑暗的地区。 I then performed a selective color adjustment , and reduced the amount of yellow in the yellows themselves.当时我的表现有选择性的颜色调整 ，以及减少的数额黄色，在黄自己。 Finally, I used a color balance adjustment to add more red and magenta in. This brings the colors closer into line with what I think suited the image.最后，我用了一个色彩平衡的调整，新增更多的红色和洋红色英寸这使颜色更密切的配合是什么，我认为适合的形象。
[img]http://www.missyuan.net/upimg/allimg/080801/11403416.jpg[/img]
Step 16步骤16
It's important in photo manipulations like this to add shadows and shading between layers to build up a proper sense of depth.它的重要性在照片操纵喜欢这个新增的阴影和明暗层间要建立一个适当的意识深入。 Generally, I tend to brush softly onto layers with black (or more often a dark color chosen from whatever object is casting the shadow).一般来说，我倾向于轻轻刷上黑色层（或更经常一个黑暗的颜色选择，无论从对象，是铸造的阴影） 。 Sometimes a quicker method is what I used here.有时，一个快捷的方法是什么我用在这里。
I created a new layer below the part of the frame I had just been working on, selected the outline of the frame, and filled it with black.我创建了一个新的层以下部分的框架，我只是一直在努力，选定纲要的框架，填补了它与黑色。 By the way, to select the outline of a layer with a layer mask, CNTRL + Click on the layer's thumbnail.由途径，选择纲要层与层口罩， cntrl +按一下该层的缩图。 Then hold down CNTRL + ALT + Shift and Click on the layer mask's thumbnail.然后按住cntrl +按Alt +转移，并点击层的面具的缩图。
I then blurred this layer.然后我模糊的这层。 A smallish Gaussian blur followed by another faded to 50% multiply did the job (CNTRL + Shift + F to use the fade command, possibly the most useful command in Photoshop in my opinion) . 1 smallish 高斯模糊其次是另一个褪色至50 ％乘以当时的就业（ cntrl +转移 + F到使用淡出命令，可能是最有用的命令，在Photoshop在我的意见） 。 I then used a warp transform to pull this shadow down slightly so it sat underneath the frame.然后我用了一个经变换退出这个阴影略有下降，所以坐在底下的框架。 At the bottom where I have faded the frame off, I didn't want the shadow, as this is where the frame is meant to be blending into the face.在底部的地方我已褪色的框架内起飞，我不想阴影，因为这是哪里的框架，就是要融入面对的问题。 I added a layer mask to the shadow, and softly masked off the shadow around this area.我已将一层面具，以阴影，并轻轻蒙面小康的阴影，周围的这方面的工作。
[img]http://www.missyuan.net/upimg/allimg/080801/11403417.jpg[/img]
Step 17步骤17
Now things started to move a little bit faster.现在的东西，开始向一点点速度更快。 First, I linked my frame part and shadow layers together and put them in a layer group for quick manipulation.首先，我联系我的框架的一部分，和阴影层一起，把他们在一个层组的快速操控。 I then started to play around with duplicating, resizing, rotating, and moving them around to create more complex shapes.然后我开始发挥左右，与架床叠屋，大小，旋转，移动和他们周围，以创造更多的复杂形状。 Here you can see how the face looks like four copies of one object on it.在这里您可以看到如何面对看起来像四份之一的对象。
I should note here that when producing a symmetrical image like this, it would be very easy just to do half of everything and flip it over at the end.我要在这里指出，当产生了对称的形象，这样，便很容易，只是做一半的一切和Flip它在去年底。 However, this can produce an overly perfect looking image that then seems fake.不过，这可以产生一种过于完美的形象，看，然后似乎是假的。 I do nearly all the symmetry by hand, and don't bother measuring or anything like that.我几乎所有的对称性手，不要理会测量或任何类似的表示。 Just place things where they feel right.刚才发生的事情，他们觉得正确的。
At this point, I also decided that his nose was too off center for my liking.在这一点上，我也决定，他的鼻子太偏离中心，为我喜欢。 I slightly moved it to the right using a very similar method to replacing the eye (duplicate, move, blend). i略提出它有权使用一个非常类似方法取代眼（复制，移动，共混物） 。
[img]http://www.missyuan.net/upimg/allimg/080801/11403418.jpg[/img]
Step 18步骤18
Time to add some more shapes into the mix.时间添加一些更多的形状纳入组合。 Here is another piece of the frame roughly brought in as before.这里是另一幅框架大致带来的，作为前。
I often add adjustment layers at the top of my layer stack while working to bring quick color changes to a whole picture.我常常添加调整层的顶部，我层堆叠，而工作带来的快速颜色变化的一个全貌。 It can help unify the various pieces of an image and give them a sense of cohesion.它可以帮助统一的各种件的形象，并给予他们的责任感凝聚力。 Which type of adjustment layer I use depends entirely on the image, but the most common ones I use are: Curves, Color balance, Gradient map, Photo filter, Hue/Saturation and Selective color .哪种类型的调整层的使用完全依赖于形象，但最常见的使用是： 曲线，色彩平衡，梯度地图，照片过滤器，色调/饱和度和选择性的颜色 。 For this image, I added a gradient map adjustment layer at this point using a dark blue to white gradient, set to a soft light blending mode with an opacity of about 20% .这个形象，我增加了一个梯度地图调整层在这一点上用深蓝色白梯度，设置为软轻混合模式与浑浊的20 ％左右 。 This helped add contrast as well as soften the oranges and yellows.这有助于增加对比度以及软化橙和黄色。
I added quite a few different adjustment layers during the making of this image at different times.我已将不少不同的调整层在作出这个形象在不同的时期。 If the image suddenly seems to change color or lightness subtly then that was probably when I added one.如果图像突然似乎改变颜色或亮度巧妙那可能是，当我说：一。
[img]http://www.missyuan.net/upimg/allimg/080801/11403419.jpg[/img]
Step 19步骤19
Properly extracted and duplicated this new piece is beginning to fit in nicely.适当的提取和重复，这是一块新的开始，以适应皆大欢喜。 You might be able to see where I tried to match some of the curving elements on the two bits of frame, so they appear to join up nicely.您也许能够看到我尝试，以配合部分的曲线元素对二位的框架，所以他们似乎加入了皆大欢喜。 This helps make the photo blending more seamlessly.这有助于使照片更完美地融合。
[img]http://www.missyuan.net/upimg/allimg/080801/11403420.jpg[/img]
Step 20步骤20
Now I added a third piece from the frame.现在我加入了第三条从框架。 Colors have been matched up and shadows added to make the pieces fit together nicely.颜色已配对和阴影，补充，使件，配合很好。
[img]http://www.missyuan.net/upimg/allimg/080801/11403421.jpg[/img]
Step 21步骤21
Now I was adding pieces continually, using the three different parts I had cut off the frame.现在我加入件不断，使用三个不同的部分，我曾切断框架。 All I had to do was resize, rotate, or distort them, and place them where I think looked good.所有我不得不做的是调整，旋转，或歪曲他们，和他们的地方，那里我觉得期待好。 I also decided to mask off the guy's ears at this point as they were just getting in the way.我也决定把面具小康家伙的耳朵，在这一点上，因为他们在刚刚起步的方式。
I also decided I want some subtle patterns on the skin in the same shapes as the main frame bits I've been adding.我也决定，我想了一些微妙的模式对皮肤在相同的形状为主要框架位我已经加入。 There are many ways this could be done, but I decided on using the highpass filter , combined with a blending mode.有很多方法可以这样做，但我决定对使用高通滤波器 ，再加上一个混合模式。 The highpass filter is a good way to 'neutralize' everything but the edges.该高通滤波器是一个好方法，以'中'一切，但边缘。 Basically, it will retain the high contrast edges while changing the other areas to mid level color.基本上，它会保留高对比度的优势，而改变其他领域中水平的颜色。 When you combine this with a soft light or overlay blending mode , which ignores mid level colors, you have a way of keeping the edge details.当您结合起来，这与软轻或重叠融合模式 ，忽略了中级别的颜色，你有办法保持边缘的细节。
Here is what one of the frame pieces looks like after being run through the highpass filter and desaturated (I didn't want any color tingeing on the skin either).这里是一个框架件看起来像后，正在运行的通过高通滤波器和desaturated （我不想任何颜色tingeing对皮肤要么） 。 This is the sort of detail to aim for, not too sharp, but not too blurry.这是排序的详细目标，也不要太尖锐，但不能过于模糊。 A 10 pixel or so setting worked here. 10像素或设置工作，所以在这里。
[img]http://www.missyuan.net/upimg/allimg/080801/11403422.jpg[/img]
Step 22步骤22
Here, the blending mode of the filtered piece has been changed to overlay , and the opacity dropped to about 60% .在这里，融合模式过滤片已改为叠加 ， 不透明度下降到60 ％左右 。 The result is the highlights and shadows of the shapes appear on the skin, but the areas between are left unaffected.结果是重点和阴影的形状出现在皮肤，但地区之间的是左，不会受到影响。
As with the main shapes, I repeated this for different parts of the frame, and continued to move and transform them around the face.作为与主要的形状，我重复，这为不同部分的框架，并继续提出和改造他们周围的面子。 Since these shapes are using a blending mode, it's important that they didn't overlap.由于这些形状是使用混合模式，它的重要的是，他们并没有重叠。 This would end up producing a muddy, blurred look.这最终会产生浑浊，模糊看看。
[img]http://www.missyuan.net/upimg/allimg/080801/11403423.jpg[/img]
Step 23步骤23
At this point the guy's face was bothering me, so I made a few changes.在这一点上家伙的脸是困扰我，所以我做了一些改动。 First, I roughly masked off the top of his head, as it was distracting, and I think I wanted to replace it later anyway.首先，我粗略蒙面小康上方的他的头部，因为它是分心，而且我觉得我想取代它稍后无论如何。 Then I realized the mouth needed lining up with the nose, so I slightly moved it to the right using exactly the same method as with the nose.然后我意识到口需要衬砌与鼻子，所以我稍微提出它有权使用完全相同的方法，作为与牵着鼻子走。
I then decided I didn't like the eyes either.于是我决定我不喜欢的眼睛。 They were just too dark and were sucking the life out of the picture.他们只是太暗，被吸的生命出来的图片。 In cases like this, where the eyes are actually just pure black, there is no good option but to replace them completely.在这种情况，那里的眼中，其实只是纯黑色，有没有好的选择，但要完全取代他们。
I found a suitable eye photo from my collection, roughly cut it out and stuck it in.我发现一个合适的眼睛的照片，我的收藏，大约削减它和坚持它英寸
[img]http://www.missyuan.net/upimg/allimg/080801/11403424.jpg[/img]
Step 24步骤24
To make sure it lined up properly I dropped its opacity to 50% .以确保它列队妥善我扔掉其不透明至50 ％ 。 Then I resized and moved it around until both the old and new pupils lined up.那么，我提出的大小和它靠近，直到双方旧的和新的学生列队。 This is the only part I was really interested in replacing, so that's all that needed to be done.这是唯一的一部分，我是真的有兴趣取代，使的所有需要做的。
Once the eye was lined up, I put the opacity back up to full, and added a reveal none layer mask .一旦眼睛被排队，我把浑浊的备份，以充分，并增加了一个显示无层的面具 。 I then softly brushed white onto the mask to reveal the new pupil.然后我轻轻刷上白色的面具，揭示了新的学生。
[img]http://www.missyuan.net/upimg/allimg/080801/11403425.jpg[/img]
Step 25步骤25
I repeated the process for the other eye.我重复的过程中为其他眼。 The new lightness of the eyes was much better, but the color was a bit dull, so I changed them to red.新轻的眼睛好得多，但颜色是有点无趣，所以我改变了他们的红色。 A combination of hue/saturation and color balance adjustments is a good way for hefty recoloring like this.相结合的色调/饱和度及色彩平衡的调整是一个好方法，为大幅recoloring喜欢这个。
[img]http://www.missyuan.net/upimg/allimg/080801/11403426.jpg[/img]
Step 26步骤26
Again, there was something bothering me about this guy's face.再次，有一些困扰我对这个家伙的脸。 This time it was the tip of his nose, which was still pointing off to one side.这个时候，这是提示他的鼻子，这仍然是指向小康一方。 A quick trip to the liquefy tool , and I pushed it into the center.一种快速去液化工具 ，我推到中心。
The piece I added under the nose is made from one of the same three shapes I cut from the frame.一幅我已将下，鼻子是由一个相同的三个形状i削减从框架。 However, here I mixed both the first technique (straight paste) and second (highpass filter) .不过，我在这里混合双方的第一技术（直贴）和第二届（高通滤波器） 。 To do this I first moved the shape into the right place, and then duplicated it.这样做，这是我第一次提出的形状到正确的位置，然后重复。
I performed the highpass filter procedure on the bottom layer, and then added a layer mask to the top layer. i履行高通滤波器的程序，底层，然后补充了一层面具 ，以最上层。 I then used a black to transparent gradient and dragged it onto the mask from the areas I wanted to hide towards the areas I wanted to keep.然后我用一个黑色透明的梯度把它推到面具，从地区我想隐藏对地区我想保持。 This way the shapes slowly fade from solid into the subtle tracery.这样的形状，慢慢淡出，从固体到微妙tracery 。
[img]http://www.missyuan.net/upimg/allimg/080801/11403427.jpg[/img]
Step 27步骤27
As well as adding some more subtle patterns to the nose, I now wanted to put some large areas of color on the cheeks.以及加入一些更微妙的模式，以鼻子，我现在想提出一些大面积的颜色就脸颊。 I quickly drew out a shape using the pen tool , and converted this into a selection.我很快制定出一个形状，使用钢笔工具 ，并转换到这一点的选择。 I intersected this with the face to make sure my new shapes wouldn't go over the background (CNTRL + ALT + Shift + Clicked on face's layer thumbnail) . i交叉，这与面对，以确保我的新形状不会超过背景（ cntrl +按Alt +转移 + 点击就面对的层缩图） 。
[img]http://www.missyuan.net/upimg/allimg/080801/11403428.jpg[/img]
Step 28步骤28
On a new layer above the face, I dragged a red to transparent gradient from the top left to bottom right.对一个新的层以上的面对，我拖了一个红色的透明梯度从左上角到右下角。 I then changed the blending mode of this layer to linear burn to blend it better.然后我改变了混合模式，这层线性烧伤 ，以更好地融合。 This looked a little flat so I decided to add some styling.这似乎不大的单位，所以我决定添加一些造型。
To do this I decided on layer styles; however, trying to add styles to a layer with soft edges like this new shape can produce awkward results.为此，我决定对层的风格;不过，尝试添加样式，一层软的优势，喜欢这个新的形状可以产生尴尬的结果。 Much better to use a layer styles dedicated layer .好得多使用层的作风，专用层 。 To do this I selected the shape's path I drew, and converted it to a selection again.这样做，这是我选定的形状的路径我提请，并转换到一个选择 。 I then created a new layer above my red shape and filled it with black (although it doesn't matter what color you use).然后我创建了一个新的层以上的，我红色的形状和填补它与黑色（虽然也不要紧，什么颜色您使用） 。 I then changed the layer's fill (not opacity!) to 0% .然后我改变了层的填写（不混浊！ ） 0 ％ 。 This allows layer styles to show while nothing in the layer itself actually appears.这使得层的风格，以显示而没有在该层本身其实出现。
I then added an inner shadow and outer glow to add a bit of definition to the red shape.当时我说， 党内的阴影和外层空间的辉光补充一点的定义，以红色的形状。 To mask off the bits of these new styles I didn't want, I made sure layer mask hides effects was ticked in the layer styles window, and used a layer mask like normal.掩盖小康位，这些新作风，我不想，我作出了肯定的面具隐藏层的影响，被分在该层的风格窗口，并使用了一层面具一样正常。
[img]http://www.missyuan.net/upimg/allimg/080801/11403429.jpg[/img]
Step 29步骤29
I linked these two shape layers together, and stuck them in a layer group for ease. i联系这两个形状层一起，并坚持他们在一个层组，为纾缓。 I then duplicated this group, and reflected it horizontally, before moving it over to the other cheek.然后我重复这个小组，并反映它的水平之前，移动到其他的脸颊。 At this point, I added another adjustment layer to add more blue to the shadows, as well as increase contrast.在这一点上，我又增加调整层 ，以增加更多的蓝色阴影，以及增加的反差。
[img]http://www.missyuan.net/upimg/allimg/080801/11403430.jpg[/img]
Step 30步骤30
Not much to say here, I just added more of the same shapes to the side of the head using the same blending method as for those below the nose.没有太大的说，在这里，我只是补充更多相同的形状，以一侧头部使用相同的配煤方法，至于那些下面牵着鼻子走。 I reduced the opacity of the cheek shapes a bit as well, as they were slightly too distracting.我把浑浊的脸颊形状有点以及，因为他们稍微过于分散。
[img]http://www.missyuan.net/upimg/allimg/080801/11403431.jpg[/img]
Step 31步骤31
I wanted to add some new shapes now before these ones started to get too monotonous, so it was time to bring in another photo.我想添加一些新的形状，现在在这些个案开始得到太单调了，所以这是个时间，使在另一张照片。 This is another picture frame photo from iStock .这是另一种画面的照片， istock 。 I was only interested in the very top of it, as shown below. Extraction and color fixing were identical to the first frame photo.我只是有兴趣在非常上方，如下图所示。 提取和固色相同的第一帧照片。
[img]http://www.missyuan.net/upimg/allimg/080801/11403432.jpg[/img]
Step 32步骤32
I pasted the new frame underneath the old ones, flipped it upside down, and masked off the parts I didn't need. i贴上新的框架下，旧的，它翻转倒，蒙面小康的部分，我没有需要。 Color matching was a little more complex here, but nothing a custom created gradient map couldn't fix.配色是一个复杂的多一点在这里，但没有一个自定义创建梯度地图无法修复。 The shadow I added here was small, as I wanted it to appear to be very close to the face.的阴影，我已将这里的小，正如我希望它似乎非常接近，面对。 A drop shadow layer style sufficed for this. 下降的阴影层的作风，足以为这一点。
[img]http://www.missyuan.net/upimg/allimg/080801/11403433.jpg[/img]
Step 33步骤33
I wanted this new style of frame to blend in with the first, which it didn't quite.我想这个新的作风框架，融入在与第一，它没有相当。 To help this I blended in a couple of pieces from the first frame over the top of the new one.以帮助，这点我是混合在一两件从第一帧超过顶部的新一。 I used a layer mask and a soft brush to blend them together.我用了一层面具和软擦 ，融入他们在一起。
[img]http://www.missyuan.net/upimg/allimg/080801/11403434.jpg[/img]
Step 34步骤34
At this point, I decided the whole face needed to be brought in line with the colors of the frames a bit more.在这一点上，我决定整个面对，必须使之与色彩的框架多一点。 To do this I did a hue/saturation adjustment , choosing colorize , and selecting a dark orangey/brown.为此我做了色调/饱和度调整 ，选择colorize ，并选择一个黑暗的orangey /布朗。
[img]http://www.missyuan.net/upimg/allimg/080801/11403435.jpg[/img]
Step 35步骤35
I then faded this (CNTRL + Shift + F) to 30% , as a full colorization always looks rubbish.当时我淡忘今（ cntrl +转移+ f ）至30 ％ ，作为一个全面的色总是期待垃圾。 Keeping a range of colors is important.保持了一系列的颜色是非常重要的。
[img]http://www.missyuan.net/upimg/allimg/080801/11403436.jpg[/img]
Step 36步骤36
The top of the head had been bothering me for some time by then.上方的头部已困扰我一段时间，届时。 I wanted some new shapes to add up there, something like a crown with an Eastern feeling.我想了一些新的形状，加上在那里，像官方与东方的感觉。 After a fair amount of searching, I found this image on iStock .之后，相当数量的搜索，我发现这个图像就istock 。
[img]http://www.missyuan.net/upimg/allimg/080801/11403437.jpg[/img]
Step 37步骤37
Before starting on the laborious job of extraction , I roughly comped in the new photo with a very rough extraction and color fix .出发之前就费力的工作， 提取 ，我粗略comped在新的照片，一个非常粗略的提取和颜色修补程序 。
[img]http://www.missyuan.net/upimg/allimg/080801/11403438.jpg[/img]
Step 38步骤38
Liking what I saw, I extracted the image properly, this time using the pen tool , as I wanted to preserve the nicely curved edges.喜欢我所看到的，我提取的形象，妥善，这个时候用钢笔工具 ，正如我要保存好，弯曲的优势。
[img]http://www.missyuan.net/upimg/allimg/080801/11403439.jpg[/img]
Step 39步骤39
Fixing the colors was again quite a bit more challenging than with the first frame.固定的颜色又相当多，更具挑战比与第一帧。 A good custom gradient map and repeated use of the color balance adjustment was necessary.一个良好的自定义梯度地图和重复使用的色彩平衡的调整是必要的。 Although you can't see here, this photo was not overly well focused either.虽然你不能在这里看到，这张照片是并不过分，以及重点。 To affect a heavy sharpen I ran the highpass filter over it with a setting of about 3 pixels .影响一个沉重激化我参加了高通滤波器超过它与设置约3个像素 。 Then I faded the filter with an overlay blending mode .然后我淡忘过滤器与一重叠融合的模式 。 This helped bring it back into focus.这促成回关注的焦点。
[img]http://www.missyuan.net/upimg/allimg/080801/11403440.jpg[/img]
Step 40步骤40
I also felt that the distribution of light over this new piece, which I will call the crown from now on, was too even.我也认为，分布光明战胜这一新的作品，我会请官方从现在起，过于甚至。 I wanted the focus to be more central.我想重点要更多的中央。 To help this I selected the outline of the crown and created a new layer.以帮助，这是我选定纲要的官方和创造了一个新的层。 I then created a white to black radial gradient from the top center outwards.然后我创建了一个白色黑色径向梯度从顶部的中心向外。 I then changed this layer's blending mode to soft light , and adjusted the opacity until it looked good.然后我改变了这层的混合模式，以柔和光线 ，并调整不透明度 ，直到它期待的好。
[img]http://www.missyuan.net/upimg/allimg/080801/11403441.jpg[/img]
Step 41步骤41
As with the chin piece, I wanted the new shapes present in the crown to blend more with the shapes already present.作为与下巴一块，我希望新的形状，目前在官方，融入更多与形状已经存在。 I copied parts of the crown, and moved them to near where the ears would be, behind the frame shapes. i复制的部分皇冠，并提出他们附近的耳朵会，背后的框架的形状。 I also used the highpass method on a duplicate of the crown to add some more tracery to the chin and forehead.我也用了Highpass方法对重复官方添加一些更tracery到下巴和额头。
[img]http://www.missyuan.net/upimg/allimg/080801/11403442.jpg[/img]
Step 42步骤42
To continue with this matching of shapes, I started to blend some parts of the first frame into the crown.继续与此匹配的形状，我开始共混物的某些部分的第一帧到皇冠。 I added some shapes around the edge and behind the crown to enhance its outline.我已将一些形状围绕优势和背后的冠，以提高其大纲。 I then blended some of the very first shapes (the shapes I had cut out directly) onto the front of the crown.当时我混纺一些非常第一的形状（形状，我曾削减直接）的前端，用皇冠。 I felt now that it was more in tune with the rest of the design work.我觉得现在，它更在调整与其余的设计工作。
[img]http://www.missyuan.net/upimg/allimg/080801/11403443.jpg[/img]
Step 43步骤43
It was about this time that the stubble above his lip started to bother me.这是有关这时候，该茬以上，他嘴唇开始困扰着我。 It seemed out of place.它似乎是出于地方。 To fix this, I simply repeated the method I used to reduce the puffiness below his eyes.要解决此问题，我只是重复的方法，我用，以减少浮皮下面他的眼睛是雪亮的。 This time I cloned skin from his forehead, and used a slightly higher opacity on the new layer to override the stubble.这一次我克隆皮肤从他的前额，并用了一个稍高的不透明对新层凌驾茬。
[img]http://www.missyuan.net/upimg/allimg/080801/11403444.jpg[/img]
Step 44步骤44
I felt the gap in the center of his forehead was a bit boring and wanted to add some more focus.我觉得差距在中心他的前额是有点枯燥，而且要添加一些更多的关注的焦点。 I thought a jewel would be a nice touch, so went off to find a decent photo.我以为宝石将是一个很好接触，所以到小康，以找到一个体面的照片。 I chose this one because of the lotus-like shape it has, which would continue with my vague Eastern theme.我选择了这一个，因为莲花般的形状，它已将继续与我含糊东部的主题。 Again, another iStock image.再次，另一istock形象。 I should get commission from them!我应该得到委员会，由他们！
[img]http://www.missyuan.net/upimg/allimg/080801/11403445.jpg[/img]
Step 45步骤45
I extracted the jewel using the pen tool , and pasted it into my main document. i提取的宝石，使用钢笔工具 ，并粘贴到我的主要文件。 After resizing and rotating it, I decided that it looked a bit too lopsided.之后，大小和旋转的话，我决定，它期待未免太片面。 To fix this I deleted one half of it, duplicated the remaining half, flipped it horizontally, and joined the two pieces together.要解决此问题我删除了一半，重复余下的一半，它横向翻转，并加入了两件一起。 After some color fixes and adding a shadow , it was looking better.之后，一些颜色修正 ，并加入了阴影 ，这是寻找更好的。 I also added another couple of small shapes from the first frame to fill in the gaps to the sides of the jewel.我也说，另一对夫妇的小形状从第一帧，以填补在的差距，向两岸的宝石。
[img]http://www.missyuan.net/upimg/allimg/080801/11403446.jpg[/img]
Step 46步骤46
The gap above the jewel now looked decidedly odd, so I decided to re-add the part of the crown I had masked off.差距以上的宝石，现在看来断然多，所以我决定重新添加的一部分，官方我曾蒙面的小康。 The great thing about using layer masks is that to get it back all I had to do was brush white onto the mask where I wanted to undelete stuff.伟大的事，有关使用层口罩是为了取回所有我要做的就是刷上白色面具的地方我想删除的东西。 I changed the colors of the jewel slightly by adding a radial gradient overlay layer style set to a soft light blending mode .我改变颜色的宝石，稍微增加了径向梯度覆盖层的风格设置为软轻混合模式 。 This darkened the center and lightened the edges.这黑暗的中心，减轻了优势。
[img]http://www.missyuan.net/upimg/allimg/080801/11403447.jpg[/img]
Step 47步骤47
The stud in the center of the jewel was annoying me slightly, and I wanted the whole thing to look a bit more lotus like.该种在该中心的宝石，是恼人的我略，我想整个事情看多些荷花一样。 To fix this, I simply duplicated the jewel and shrank the top copy by 50% .要解决此问题，我只是重复的宝石和减少顶端复制了50 ％ 。 Then I added a tight drop shadow layer style to this copy to make it stand out a bit more.然后我增添了紧张的下拉阴影层的作风，这个复制，使其站出来多一点。
[img]http://www.missyuan.net/upimg/allimg/080801/11403448.jpg[/img]
Step 48步骤48
I felt that I was getting near the end of the image now so started dealing with small tweaks.我觉得我是越来越接近尾声的形象，现在开始处理小的调整。 I added some tracery to the cheeks using the same highpass method .我已将一些tracery ，以脸颊使用相同的了Highpass方法 。 I cloned out the blob on the very bottom of the jewel, and added a couple of dashed lines above each eye. i 克隆出的BLOB就非常底部的宝石，并补充说：一对夫妇的虚线以上的每只眼睛。 These were actually imported from Illustrator as dashed strokes are one of its default options.这些人其实是从外地输入Illustrator中，作为落空，中风是其一个预设选项。 For people without Illustrator I have another very old tutorial that can help replicate the effect in Photoshop: Dashed stroke .为人民没有Illustrator中我还有一个很旧的补习，可以帮助复制的效果在Photoshop ： 冲出中风 。
[img]http://www.missyuan.net/upimg/allimg/080801/11403449.jpg[/img]
Step 49步骤49
For some final details, I wanted to add some actual petals in. I found this photo of a lotus, although since these were going to be very small details, I didn't look too hard.一些最后细节，我想补充一些实际的花瓣英寸我发现这张照片的荷花，虽然因为这些将会非常小的细节，我不看太辛苦。 Any old thing would have done here.任何旧事，这样做在这里。
[img]http://www.missyuan.net/upimg/allimg/080801/11403450.jpg[/img]
Step 50步骤50
I extracted the flower, darkened it considerably using a curves adjustment , and added a fair amount of yellow and red using a color balance adjustment . i提取花，黑暗的，它相当使用曲线调整 ，并增添了相当数量的黄色和红色的使用色彩平衡的调整 。 This was to make sure the petals didn't stand out too much, and that they blended in with the rest of the design.这是为了确保花瓣没有站出来太多了，和他们混在与其余的设计。 A simple drop shadow layer style was fine for the shadows.一个简单的下拉阴影层的作风是罚款的阴影。 I then placed the flower between the face and the frame layers, duplicated, transformed, and moved them a few times into various positions.然后我把花之间的面部和框架层，复制，转化，并提出他们的几倍到各种立场。
[img]http://www.missyuan.net/upimg/allimg/080801/11403451.jpg[/img]
Step 51步骤51
The jewel had been standing out too much for me at this point, so I darkened it heavily with a gradient map adjustment that goes from dark blue to white and faded to a soft light blending mode and low opacity .宝石，已站在了太多，我在这一点上，所以我黑暗的，它与大量的梯度地图的调整 ，从深蓝色白褪色，以软轻混合模式和不透明度低 。 I then boosted the reds slightly using a selective hue/saturation adjustment , which defined the exact range of colors to affect using the sliders at the bottom.当时我增强了红人队稍微使用选择性的色调/饱和度调整 ，其中界定的确切范围的颜色，影响使用滑块在底部。
[img]http://www.missyuan.net/upimg/allimg/080801/11403452.jpg[/img]
Step 52步骤52
I felt that the image was nearly done by now, so decided to start work on the background.我觉得形象，几乎是现在所做的，所以决定开始工作的背景。 I wanted this to be very minimal so as to not distract at all from what I had done so far.我想这是很小，以便不使分散于各从我已经这样做。 I decided I wanted a starburst style effect, but rather than use any of the normal techniques I decided to experiment and come up with something newer.我决定我想一starburst作风的影响，而是比使用任何正常的技术，我决定把实验和想出了一些较新。 The first thing I needed to do was to get a flat copy of the image, so I flattened it, and copied this to a new document (making sure to undo the flattening to the main picture afterwards).第一件事，我需要做的是为了获得一个单位的副本的形象，所以我夷为平地，并且复制，这到一个新的文件（请务必撤消平缓，以主要图片之后） 。
I had an idea that involved isolating the details of the image and then using them to produce a burst.我有一个想法，涉及孤立的细节图像，然后利用他们产生了一阵。 As before, a highpass was the best way to isolate these details.一如以往，一了Highpass是最好的方式，来孤立这些细节问题。
[img]http://www.missyuan.net/upimg/allimg/080801/11403453.jpg[/img]
Step 53步骤53
I was trying to produce a selection based around just the edges of the image, so I needed to copy this filtered image into a new channel .我是试图产生一个选择基于周围的只是边缘的形象，所以我需要复制此过滤的形象，进入了一个新渠道 。 The next step was to make sure that most of the image wouldn't be selected, ie it would be black in my channel.下一步是要确保大部分的形象会不会被选择，即它会在我的黑色通道。 Using the set black point dropper in the curves palette, I moved everything except the lightest shades to black.使用一套黑点滴管在曲线调色盘，我提出的一切，除最轻的色调以黑色。 I then smoothed some of the noise out by running it through a low setting median filter , and then reset the white point as the median had reduced the contrast too much.然后我平滑的一些噪音是由运行它通过一个低设置中值滤波 ，然后重置白点为中位数减少了反差太大。
[img]http://www.missyuan.net/upimg/allimg/080801/11403454.jpg[/img]
Step 54步骤54
This is where I used a function of Photoshop that is rarely touched on, but can be very useful in certain circumstances.这是我用一个函数的Photoshop是很少触及，但可以非常有用，在某些情况下。 I got a selection from my custom channel , and then used the Make work path from selection function in the paths palette.我得到的选择从我的自订频道 ，然后用，使工作路径选择功能，在路径调色盘。 I now had a very complex path based around the high contrast areas of the image.我现在有一个非常复杂的路径为基础的周围，高对比度领域的形象。
On a new layer I then stroked this path with a normal round hard brush , making sure that spacing was set to 1% , and simulate pressure was ticked in the stroke sub menu.对一个新的层，我当时抚摸这条道路，与正常的轮硬刷子 ，确保间距设置为1 ％ ，和模拟的压力是分在中风次级菜单。 I now had many squiggly shapes on this new layer.我现在有许多弯曲的形状对这一新的层。 Fine so far but nothing like a starburst.罚款直至目前为止，但没有像starburst 。
[img]http://www.missyuan.net/upimg/allimg/080801/11403455.jpg[/img]
Step 55步骤55
To turn this into more interesting shapes I needed to duplicate this layer a number of times, while increasing the size slightly each time.把到这一点更有趣的形状，我需要重复的这层了很多次，而增加人数略每一次。 The easiest way to do repetitive changes like this is to use a custom action .最简单的方式，做重复的变化一样，这是使用自定义的行动 。 I recorded an action which duplicated the current layer, increased its size to 100.5% , and decreased its lightness slightly.纪录一行动，重复当前层，增加其大小为100.5 ％ ，并降低其明度稍微。 I then hit the play button on this action about 25 times to produce the starburst.当时我击中播放按钮就这项行动约25倍，产生starburst 。
To be honest there are many ways this effect could be produced, this is just what I came up with on the spur of the moment.以诚实有很多方法的效果，这可以产生，这只是我想出了对一时冲动。 This is a bit more interesting than your usual radial blur/polar coordinates methods though.这是一个较为有趣的比您平常径向模糊/极坐标的方法，虽然。
[img]http://www.missyuan.net/upimg/allimg/080801/11403456.jpg[/img]
Step 56步骤56
I stamped all these duplicated layers into one (select all layers, CNTRL + ALT + E) , and copied and pasted back into my main document just above the background layer. i盖上所有这些重复层到1 （选择所有图层， cntrl +按Alt + E ）和复制并粘贴回我的主要文件，只是上述的背景层。 I increased its size slightly, so it wasn't completely hidden, and roughly aligned it with the head. i增加其人数略，因此无法完全隐藏，并大致不结盟，它与头部。
[img]http://www.missyuan.net/upimg/allimg/080801/11403457.jpg[/img]
Step 57步骤57
To add a bit of interest into the background, I introduced some elements from a new photo, this time from Stock.xchng .添加一点利益的背景下，我介绍的一些内容，从一个新的照片，这个时候从stock.xchng 。 After setting the black point darker , so only the bright sparks were visible, I desaturated it slightly, and changed its blending mode to linear dodge .后设置的黑点较深 ，因此只有光明的火花，可见，我desaturated它略，并改变其混合模式线性回避 。
[img]http://www.missyuan.net/upimg/allimg/080801/11403458.jpg[/img]
Step 58步骤58
I then placed a few different copies around the image, so that some sparks shot out from different parts behind the head.然后我放在几个不同的副本，围绕形象，使一些火花，拍摄出来的不同部分背后的头部。 I masked off any sharp edges using a normal layer mask . i蒙面小康任何锋利使用正常层的面具 。 I used a soft brush on another layer with a screen blending mode to add a bit more light.我用软毛笔在另一个层与屏幕混合模式 ，以添加一个较为轻。 Then I placed a color balance adjustment layer above all of these new background layers to add some more magenta into the mix.那么，我放置了一个色彩平衡的调整层以上的所有这些新的背景层添加一些更多的洋红纳入组合。
[img]http://www.missyuan.net/upimg/allimg/080801/11403459.jpg[/img]
Step 59步骤59
At this point I was definitely fed up with the image and wanted to get it finished; So, I made the final few tweaks before flattening down.在这一点上我是绝对厌倦了的形象，并希望得到它完成;所以，我作出最后几年的调整之前，平缓下来。 I added a few tiny skulls into the crown to cover up some relatively blank areas, and messed about the with lighting distribution of the background, nothing major.我已将一些微小的头骨到官方掩盖了一些相对空白的领域，和搞砸有关照明分布的背景，没有什么重大。
[img]http://www.missyuan.net/upimg/allimg/080801/11403460.jpg[/img]
Step 60步骤60
Flattening down at the end of an image is an important step, and not just for saving a flattened version.平坦的下降在去年底的形象是一个重要的步骤，而不是只为救人一夷为平地的版本。 It allows you to affect changes that would be almost impossible (not to mention incredibly time-consuming) on a layered file.它允许你影响的变化，这将几乎是不可能的（更不用提令人难以置信的时间消耗）对分层文件。 How much or little you do on any particular image will vary hugely depending on what effect you are after.多少或很少这样做，你就任何特定的形象，会有所不同，视乎巨大什么影响你之后。 I will describe what I did for this image though.我会说明我所做的这个形象，虽然。
After flattening and copying the image into a new document (always best to do this just in case you accidentally save over your layered file), I first zoomed in and looked for any small details that needed fixing.后扁平化和复制的形象，进入了一个新的文件（最好是这样做只是在情况下，您意外地保存了您的分层文件） ，我第一次在缩放，并期待为任何一个小细节需要确定。 The clone and smudge tools can be used in cases like this to quickly fix any small problems, jagged edges, or the like. 克隆与黑灰的工具可以用来在这种情况迅速修正任何小问题，锯齿状边缘，或类似的。 In this case, I needed to do very little.在这种情况下，我需要做的很少。
Next came the lighting and color fixes.明年而来的光线和色彩修正。 First, I ran an auto-levels adjustment on the whole image, and faded this to 50% .首先，我然自动水平调整对整个图像，褪色，这50 ％ 。 This helped unify the colors, removing some of the yellow tinge, and increasing the contrast slightly.这有助于统一的颜色，去除一些黄色色彩，和增加的对比略。 The next step involved blurring, so I first ran a light unsharp mask (30, 1.0, 0 is good for an image this size) on it first.下一步所涉及的模糊，所以我首先然轻锐化掩模（ 30岁， 1.0 ， 0是一个良好的形象，这大小）对第一次。
I then ran a lens blur over the whole image with a setting of about 30 pixels .我当时冉镜头模糊超过整个图象与设置约30个像素 。 If you can't be bothered to wait for a lens blur in cases like this, as it can take ages, a Gaussian or box blur is fine.如果您不能困扰等待一个镜头模糊，在这样的案件，因为它可以采取的年龄， 高斯模糊或方块是好的。 I then faded this to soft light blending mode with an opacity of about 30% .当时我淡忘这柔和光线混合模式与浑浊的30 ％左右 。 This gives the whole image a softer, more ethereal feel, as well as giving the contrast a bit of a dramatic boost.这使整个图像更柔和，更醚的感受，以及给予的对比有点急剧增加。
Next I added some soft highlights using a similar method.接下来，我增加了一些软性突出使用了相似的方法。 First, I duplicated my image.首先，我重复我的形象。 Then I ran another lens blur on it, but this time faded to a linear dodge blending mode with an opacity of about 15% .然后我参加另一个镜头就模糊 ，但这个时候褪色为线性道奇混合模式与浑浊的约15 ％ 。 I didn't want these highlights everywhere, so I then added a reveal none layer mask .我不想这些突出到处，所以我，然后添加了显示无层的面具 。 Using a white to transparent radial gradient and soft brush , I then added white to the mask where I wanted the image to shine a bit more.用白色透明的径向梯度和软擦 ，然后我补充说：白色面具的地方我想的形象，焕发出多一点。 In this case, I wanted it to shine around the eyes and center of the face.在这种情况下，我希望它照耀周围的眼睛和中心面对。
That's pretty much it.说的相当。 All I needed to do then was run another unsharp mask to bring some of the details back out that all that blurring had dulled.所有我需要做的，当时运行另一个锐化的面具 ，使一些细节回到指出，一切模糊了dulled 。 To produce a web resolution version, I simply shrank the image down and ran another unsharp mask , but making sure the radius was set to 0.6 pixels, rather than 1 .制作网页的决议版本，我只下跌的形象和冉另一个锐化的面具 ，但确保半径设置为0.6像素，而不是1 。
[img]http://www.missyuan.net/upimg/allimg/080801/11403461.jpg[/img]
Conclusion结论
So there we have it.因此，我们有。 I've run out of things to say.我已经一发不可收拾的东西的话。 I hope some of it made sense anyway.我希望一些它有道理的，无论如何。 Have fun!有好玩！
[img]http://www.missyuan.net/upimg/allimg/080801/1140340.jpg[/img]