【正文】 ty values in an oval region inside the window. Pixels outside the oval (shown in Fig. 2a) may represent the background, so those intensity values are ignored in puting the lighting variation across the face. The linear function will approximate the overall brightness of each part of the window, and can be subtracted from the window to pensate for a variety of lighting conditions. Then histogram equalization is performed, which nonlinearly maps the intensity values to expand the range of intensities in the window. The histogram is puted for pixels inside an oval region in the window. This pensates for differences in camera input gains, as well as improving contrast in some cases. The preprocessing steps are shown in Fig. 2.The preprocessed window is then passed through a neural network. The network has retinalconnections to its input layer。 the receptive fields of hidden units are shown in Fig. 1. There are three types of hidden units: 4 which look at 10x10 pixel subregions, 16 which look at 5x5 pixel subregions, and 6 which look at overlapping 20x5 pixel horizontal stripes of pixels. Each of these types was chosen to allow the hidden units to detect local features that might be important forface detection. In particular, the horizontal stripes allow the hidden units to detect such features as mouths or pairs of eyes, while the hidden units with square receptive fields might detect features such as individual eyes, the nose, or corners of the mouth. Although the figure shows a single hidden unit for each subregion of the input, these units can be replicated. For the experiments which are described later, we use networks with two and three sets of these hidden units. Similar input connection patterns are monly used in speech and character recognition tasks [10, 24]. The network has a single, realvalued output, which indicates whether or not the window contains a face.Examples of output from a single network are shown in Fig. 3. In the figure, each box representsthe position and size of a window to which the neural network gave a positive response. The network has some invariance to position and scale, which results in multiple boxes around some faces. Note also that there are some false detections。 they will be eliminated by methods presented in Section .TTo train the neural network used in stage one to serve as an accurate filter, a large number offace and nonface images are needed. Nearly 1050 face examples were gathered from face databases at CMU, Harvard2, and from the World Wide Web. The images contained faces of various sizes, orientations, positions, and intensities. The eyes, tip of nose, and corners and center of the mouth of each face were labelled manually. hese points were used to normalize each face to the same scale, orientation, and position, as follows:F11. Initialize , a vector which will be the average positions of each labelled feature over all the faces, with the feature locations in the first face F.F2. The feature coordinates in are rotated, translated, and scaled, so that the average locations of the eyes will appear at predetermined locations in a 20x20 pixel window.3. For each face i, pute the best rotation, translation, and scaling to align the face’s featuresFwith the average feature locations. Such transformations can be written as a linearfunction of their parameters. Thus, we can write a system of linear equations mapping thefeatures from Fto. The least squares solution to this overconstrained system yields thei0parameters for the best alignment transformation. Call the aligned feature locations F.4. Update by averaging the aligned feature locationsfor each face .5. Go to step 2.The alignment algorithm converges within five iterations, yielding for each face a function whichmaps that face to a 20x20 pixel window. Fifteen face examples are generated for the training set from each original image, by randomly rotating the images (about their center points) up to 10,scaling between 90% and 110%, translating up to half a pixel, and mirroring. Each 20x20 window in the set is then preprocessed (by applying lighting correction and histogram equalization). A few example images are shown in Fig. 4. The randomization gives the filter invariance to translations of less than a pixel and scalings of 20%. Larger changes in translation and scale are dealt with by applying the filter at every pixel position in an image pyramid, in which the images are scaled by factors of .Practically any image can serve as a nonface example because the space of nonface images ismuch larger than the space of face images. However, collecting a “representative” set of nonfacesis difficult. Instead of collecting the images before training is started, the images are collected during training, in the following manner, adapted from [21]:1. Create an initial set of nonface images by generating 1000 random images. Apply the pre processing steps to each of these images.2. Train a neural network to produce an output of 1 for the face examples, and 1 for the nonface examples. The training algorithm is standard error backpropogation with momentum [8]. On the first iteration of this loop, the network’s weights are initialized randomly. After the first iteration, we use the weights puted by training in the previous iteration as the starting point.3. Run the system on an image of scenery which contains no faces. Collect subimages in which the network incorrectly identifies a face (an output activation ).4. Select up to 250 of these subimages at random, apply the preprocessing steps, and add them into the training set as negative