为了实现高速目标的快速识别及定位, 采用多光谱辐射与声定位技术联用, 在被动探测的状态下快速获取被测目标爆炸波的光谱特性及声特性, 从而分析判断其种类、 位置等重要参数信息。 推导了多光谱辐射目标识别检出公式, 计算了采用激波声定位方法获取光程距离的精度及合理的积分时间。 实验分别采用5.56 mm北约弹、 7.62 mm 56式步枪弹、 12.7 mm 54式机枪弹作为被识别目标弹丸, 通过丹麦B&K公司的2209脉冲声压计完成声压峰值采集, 获得了三种目标弹丸的速度、 距离信息。 由静态傅里叶变换干涉系统与ICX387AL型CCD完成干涉条纹的采集, 再将获得的光程信息代入多光谱目标识别算法中, 从而通过目标的光谱特性识别弹丸的类型。 实验结果显示, 5.56 mm北约弹的三个特征波长位置幅值接近, 966 nm位置处幅值最大; 7.62 mm步枪弹在935 nm位置上幅值最大, 其他两个位置上幅值相近; 12.7 mm机枪弹的三个波长呈阶梯状分布。 在多次测试过程中, 经光谱标定、 特征波长提取等方法对采集得到的三组光谱数据均可成功识别弹丸的种类并给出目标位置, 满足设计要求。

In order to achieve quick recognition and positioning of the high-speed target, using multi-spectral radiation combined with acoustic positioning technology, in the passive state, the blast wave spectral characteristics and acoustic characteristics of the measured target were rapidly obtained, thus analysis was performed to determine the type, location and other important parameters. Multi-spectral radiation detection target recognition formula was deduced. The accuracy of the optical path length and the logical integration time was calculated by shock acoustic positioning method. Experiments used 5.56 mm NATO bullets, 7.62 mm 56-rifle bullets, 12.7 mm 54 type machine-gun bullets as a target identified projectile. Interference fringes were collected by the static Fourier transform interferometer system and ICX387AL type CCD, and the peak of sound pressure was collected using 2209 pulse sound pressure meter made by B & K Company from Denmark. Experimental results show that for the 5.56 mm NATO bullets, the three characteristic wavelengths position amplitudes are close to each other, with the maximum amplitude at 966 nm; For the 7.62 mm 56-rifle bullets, 935 nm is the maximum amplitude position, while for 966 and 997 nm position the magnitudes are sunukar; For 12.7 mm 54 type machine-gun bullets, the three wavelengths show a ladder-like distribution. With the increase in the detection distance spectral radiation energy decreased. Meanwhile, with the decrease in the total radiation spectrum, the spectrum of target was affected strongly by background noise, and the SNR of system was decreased. But the spectral characteristics of different target still exist, the target species can be identified by the system with the ratio algorithm of characteristic peaks. Through spectral calibration and characteristic wavelengths extraction, the target can successfully identify the type of projectile and target position, and it meets the design requirements.