Embedded Files
[See English Translation at the end]
看了《最强大脑》中国队的郑才千对意大利队的弗兰克的魔方墙找茬后，发现原来用看立体图的方法是可以很容易找出两张密密麻麻的图片的不同之处的。例如下图中有３个色块颜色不同：
其视图方法如下：
如果你是高手，可以做到“左眼看左图，右眼看右图”，那么就用平行视图法；否则一般用交叉视图法，即“左眼看右图，右眼看左图”：从左图中心到你右眼画一条直线，再从右图中心到你左眼画一条直线，你对着这两条线的交点看（可以借助小物体辅助），但是把焦距拉到图片位置，而不是交点位置。这时你会把这两张图片看成左中右三张图片，盯着中间那张图看，找出像激光防伪商标一样闪晶晶的３个色块（它们的大概位置在左中、左上角和右上角）。
其物理原理如下：
在真实世界中，因为人的两眼位置不同，所以看到的图片也略有不同。人的大脑能够通过分析左右眼图像的差异来辨别物体的表面材料特征（material property）以及物体的空间位置。当人的两只眼睛看到两张几乎完全相同的图像时，其中相异的区域会被大脑识别成：具有金属光泽的光滑表面 或是 激光防伪商标表面。因为这两种表面在反射光线上具有很高的各向异性(anisotropy)，即：从不同角度、不同位置看差别很大。例如：光滑表面遵循镜面反射原理，反射光方向高度取决于入射角度，所以两眼所看到的图像会因为眼睛位置不同、看的方向不同而相差很大。
所以只要学会利用大脑的这个视觉信息处理能力，我们就能很容易的找出两张图片的不同之处了，例如：像 iphone/ipad 上的“大家来找茬”之类游戏可以轻松玩通所有关。不过如果想要在像上图这样高密度色块中找茬，则仍需稍加练习一番。
My English Translation is poor:
This refers to the TV series "Super Brain" in the session between Zheng Caiqian from China and Franco from Italy: spotting the difference between two walls of Rubix Cubes. It is found that they use the method of "autostereogram vision" to spot the differences between the two images of super-crowded square dots. For example, in the above two images, there are only three square dots which have different colors.
The viewing method is as follows:
If you are a master in viewing autostereograms, you can either use parallel-view (i.e., use your left eye to look at the left image and use your right eye to look at the right image), or use the cross-view method (i.e., use your left eye to look at the right image and use your right eye to look at the left image). Most people use the cross-view method. It can be done in the following way, draw a straight line from the center of the left image to your right eye, draw another line from the center of the right image to your left eye, look towards the intersection of the two lines (you can use some small objects to assist your eyes), but pull the focal length to the image location instead of the intersection point. Then you will see these two images as three images: left, middle, right; stare at the middle image, try to find the three flashing-colored pixels which appears to be laser anti-counterfeit trademarks (their approximate locations are center-left, left-top, and right-top).
The physical principle is as follows:
In the physical world, a person's two eyes have slightly different spatial positions, therefore they see slightly different images. The human brain is evolved to be able to analyze the differences between the left-eye image and right-eye image in order to identify the material property of an object's surface and to estimate the spatial distance of the object from the eye. When a person's two eyes see two nearly identical images, the brain will interpret the region of difference as: a smooth and reflective surface with metallic lustre or a laser trademark's surface. This is because these two kinds of surfaces are highly anisotropic in reflecting light, i.e., lights reflected from different angles and positions look very different. For example: a smooth/lustrous surface obeys the principle of specular reflection, the reflected light is highly dependent on the direction of the angle of incidence. Since the two eyes have different spatial positions and they are looking from slightly different directions. Thus, the two images seen by the two eyes will differ greatly because
Therefore, if we make use of our brain's visual information processing capability, we can easily spot the differences between two nearly identical images, for example: like the iphone/ipad game, "FindIt", you can easily pass all the stages using the above method. However, if you want to spot the differences in the kinds of images as shown above which consist of so many tiny pixels, you might need a little more practice.
Page updated
Google Sites
Report abuse