【導(dǎo)讀】投影曲線為特征，采用小波邊緣檢測和搜索式無監(jiān)督聚類，結(jié)合視覺幾何不變性，實(shí)現(xiàn)亞像素級的立體精確匹配；并采用小波多尺度多分辨率的特性，拼接圖像，實(shí)驗(yàn)表明，該技術(shù)設(shè)備簡單，測量速度快，測量精度控制在mm/m以內(nèi)。目前，三維測量仍以三維坐標(biāo)測量機(jī)為主。但是它由于體積大、結(jié)構(gòu)復(fù)雜而。不能在線測量，是接觸測量而不能測量柔軟的物體。因此，研究快速無損、非接。觸在線測量在工業(yè)上十分重要。因此，在本文使用雙攝像機(jī)融合光學(xué)軸抓拍物體。應(yīng)用于測量系統(tǒng)中的多點(diǎn)壓成型機(jī)的測量，并取得了良好的效果。一般來說，有兩個(gè)問題影響的三維物體獲得確切的消息：一。經(jīng)網(wǎng)絡(luò)是用來映射關(guān)系接近的情況下攝像機(jī)標(biāo)定。然而，有一些物體的表面很光滑。匹配功能不明顯，因此用光柵對象預(yù)測。測量系統(tǒng)的測量范圍。標(biāo)與攝像機(jī)之間的兩個(gè)基準(zhǔn)距離控制三維測量系統(tǒng)的測量范圍。，所以群體總數(shù)減少；

　　

【正文】 there are two geometric invariants whether mode of geometric projection is changed. 113 4 1 21 ( ) ( )tr P P tr P PI ???? (8) 114 3 2 12 ( ) ( )tr P P tr P PI ???? (9) In this way, straight lines and curves are matched effectively. 16 In this paper, grating is projected on object in vertical direction and lateral direction respectively, while two cameras grab images. With wavelet edge detection, searching nonsupervisor, edge points are fitted into quadratic curves. With geometric invariance, quadratic curves are matched, and cross points of vertical curves and lateral curves are puted. So stereo precise matching at subpixel level is achieved. 5 Image Mosaic Based on Wavelet When largescale surface is measured, many pairs of stereo images are grabbed from different viewpoints or with moving and rotating object. And two adjacent images need mosaic. The important question of image mosaic is image registration, that is to say, overlapped parts between two adjacent images are put in order, and corresponding matching between two adjacent images is also involved in image mosaic. Corresponding matching in registration is deferred from corresponding matching in stereo vision. In the former, two images are grabbed from the same viewpoint or from the different viewpoints that are translated along the basic line, and in the latter, the viewpoints of two images are not only translated but also revolved. In this paper, some black points are pasted at random on object in order to mosaic easily, and the black points are regarded as registration feature points. Meanwhile, we use biorthogonal wavelet with linearity and symmetry to depose two images that are will be mosaic, so the images can be matched and registered from coarse to fine on multiscale, and lastly a big image is bee. In fact, wavelet transform is a bandpass filter, wavelet vector on different scales shows the stated width of frequency band, and so frequency bandwidth of each wavelet vector is unequal. Two images are deposed into wavelet vectors on different frequency bands based on Mallat algorithm [6,7] , and then the different mosaic widths are selected on different scales to match and register, so a big mosaic image is smooth and fine. 6 Experiments and Results Analysis In this paper, the technology is applied to the measurement system of MPF machine. After the shaped surface is measured, the measuring results are returned to CAD/CAE system, and closedloop control is achieved. According to measurement 17 condition, measurement precision and once shaped area, we select two cameras (MTV1881CB), two camera lenses and a imagerecord device (METEOR). The distance between two cameras is 300mm。 the distance between the surface and the two cameras is 500mm. A 150150 mm 2 curve surface is measured with the technology, measurement results are shown in Tab. 1, and measurement steps are given as follows: 1) Establish the mapping relation between image points and special points。 2) 3D surface is carried on the worktable, firstly, two cameras grab synchronously a pair of stereo images without projecting and with light. Secondly, grab two pairs of stereo images, the one is grabbed while grating is put on vertical orientation, the other is grabbed while grating is put in lateral direction。 3) Process image, remove background, and reduce noise, just as , 3b。 4) Detect feature, just as 。 5) Search corresponding points, and mosaic image。 6) Calculate 3D coordinate, reconstruct 3D surface, as shown in . Experiment shows measurement error is less than , measured time is about 2s including grabbing image, processing image, establishing the mapping relation of image point and special point, searching corresponding points and puting coordinate. 18 7 Conclusions In this paper, a new 3D measurement technology based on puter vision is proposed with simple equipment, fast measurement speed and low cost. Large object can be measured, and measurement precision is less than mm/m. It also provides a new idea for puter vision applied to industry. Experiments show that the 3D measurement technology has good robustness.