[{"data":1,"prerenderedAt":605},["ShallowReactive",2],{"content-query-tLiqH5v23B":3},{"_path":4,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":8,"description":9,"date":10,"cover":11,"type":12,"category":13,"body":14,"_type":599,"_id":600,"_source":601,"_file":602,"_stem":603,"_extension":604},"/technology-blogs/zh/3580","zh",false,"","基于MindSpore香橙派AIpro实现垃圾回收AI识别方案三：基于MobileNetv2的垃圾分类实践","作者：完美记号   原文链接：https://www.hiascend.com/developer/blog/details/0272157975874351541","2025-01-14","https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/01/17/a03f27ba6ecd4003804d59552a1be8dd.png","technology-blogs","开发者分享",{"type":15,"children":16,"toc":596},"root",[17,25,37,53,58,63,71,76,96,104,109,127,135,140,147,152,170,177,182,189,197,202,207,214,221,228,237,245,250,257,280,285,298,305,313,318,328,333,340,348,353,361,366,374,379,387,392,400,408,413,418,423,430,435,440,448,453,460,465,470,475,480,485,490,495,500,505,510,518,526,531,538,543,551,558,563,571,579,584,591],{"type":18,"tag":19,"props":20,"children":22},"element","h1",{"id":21},"基于mindspore香橙派aipro实现垃圾回收ai识别方案三基于mobilenetv2的垃圾分类实践",[23],{"type":24,"value":8},"text",{"type":18,"tag":26,"props":27,"children":28},"p",{},[29,31],{"type":24,"value":30},"**作****者：**",{"type":18,"tag":32,"props":33,"children":34},"strong",{},[35],{"type":24,"value":36},"完美记号",{"type":18,"tag":26,"props":38,"children":39},{},[40,42],{"type":24,"value":41},"**原文链接：**",{"type":18,"tag":32,"props":43,"children":44},{},[45],{"type":18,"tag":46,"props":47,"children":51},"a",{"href":48,"rel":49},"https://www.hiascend.com/developer/blog/details/0272157975874351541",[50],"nofollow",[52],{"type":24,"value":48},{"type":18,"tag":26,"props":54,"children":55},{},[56],{"type":24,"value":57},"昇腾样例仓提供了一系列媒体数据处理（DVPP/AIPP）、算子开发与调用（Ascend C）、推理应用开发与部署（AscendCL）等场景的丰富代码样例给开发者进行参考学习，帮助开发者快速入门，进而熟练掌握CANN关键特性使用。",{"type":18,"tag":26,"props":59,"children":60},{},[61],{"type":24,"value":62},"通过官方的示例可了解到基于昇腾软件栈的MobileNetV2垃圾分类的训练过程，通过使用mobilenetV2模型对输入图片进行分类推理，提供待推理的jpg图片，进行推理后，输出推理后的jpg图片。",{"type":18,"tag":26,"props":64,"children":65},{},[66],{"type":18,"tag":32,"props":67,"children":68},{},[69],{"type":24,"value":70},"OrangePi AIpro的亮点",{"type":18,"tag":26,"props":72,"children":73},{},[74],{"type":24,"value":75},"OrangePi AIpro(8-12T)采用昇腾AI技术路线，具体为4核64位处理器+AI处理器，集成图形处理器，支持8-12TOPS AI算力，拥有8GB/16GB LPDDR4X，可以外接32GB/64GB/128GB/256GB eMMC模块，支持双4K高清输出。",{"type":18,"tag":77,"props":78,"children":79},"ul",{},[80,86,91],{"type":18,"tag":81,"props":82,"children":83},"li",{},[84],{"type":24,"value":85},"丰富的接口：包括两个HDMI输出、GPIO接口、Type-C电源接口、支持SATA/NVMe SSD 2280的M.2插槽、TF插槽、千兆网口、两个USB3.0、一个USB Type-C 3.0、一个Micro USB（串口打印调试功能）、两个MIPI摄像头、一个MIPI屏等，预留电池接口。",{"type":18,"tag":81,"props":87,"children":88},{},[89],{"type":24,"value":90},"广泛适的应用场景：可用于AI边缘计算、深度视觉学习及视频流AI分析、视频图像分析、自然语言处理、智能小车、机械臂、人工智能、无人机、云计算、AR/VR、智能安防、智能家居等领域，覆盖 AIoT各个行业。",{"type":18,"tag":81,"props":92,"children":93},{},[94],{"type":24,"value":95},"系统支持：Orange Pi AIpro支持Ubuntu、openEuler操作系统，满足大多数AI算法原型验证、推理应用开发的需求。",{"type":18,"tag":26,"props":97,"children":98},{},[99],{"type":18,"tag":32,"props":100,"children":101},{},[102],{"type":24,"value":103},"环境搭建",{"type":18,"tag":26,"props":105,"children":106},{},[107],{"type":24,"value":108},"在评测之前，除了官方提供了一些配件外，还有一些需要自行提前准备的东西，自行提供的配件清单：",{"type":18,"tag":77,"props":110,"children":111},{},[112,117,122],{"type":18,"tag":81,"props":113,"children":114},{},[115],{"type":24,"value":116},"TF读卡器",{"type":18,"tag":81,"props":118,"children":119},{},[120],{"type":24,"value":121},"HDMI显示器连接线",{"type":18,"tag":81,"props":123,"children":124},{},[125],{"type":24,"value":126},"串口测试线",{"type":18,"tag":26,"props":128,"children":129},{},[130],{"type":18,"tag":131,"props":132,"children":134},"img",{"alt":7,"src":133},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/01/17/c9c827c3d1924298bc1d7827b9f8dfaf.png",[],{"type":18,"tag":26,"props":136,"children":137},{},[138],{"type":24,"value":139},"首先打开balena烧录工具地址，因为个人办公是Mac电脑，借用了同事的Windows电脑来进行此次的测试，我们选择OS为“WINDOWS”的X86|X64版本下载。",{"type":18,"tag":26,"props":141,"children":142},{},[143],{"type":18,"tag":131,"props":144,"children":146},{"alt":7,"src":145},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/01/17/545c29ca97df424496b0247321e5ce1a.png",[],{"type":18,"tag":26,"props":148,"children":149},{},[150],{"type":24,"value":151},"打开下载的balena烧录工具，不用安装，直接就打开了，非常的方便，不用在电脑上安装软件：",{"type":18,"tag":77,"props":153,"children":154},{},[155,160,165],{"type":18,"tag":81,"props":156,"children":157},{},[158],{"type":24,"value":159},"点击“从文件烧录”选择好我们提前下载的镜像（百度云真的慢）",{"type":18,"tag":81,"props":161,"children":162},{},[163],{"type":24,"value":164},"点击“选择目标磁盘”选中我们插入的TF卡(我提前格式化了)",{"type":18,"tag":81,"props":166,"children":167},{},[168],{"type":24,"value":169},"点击“现在烧录”就可以开始进行烧录了",{"type":18,"tag":26,"props":171,"children":172},{},[173],{"type":18,"tag":131,"props":174,"children":176},{"alt":7,"src":175},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/01/17/8adde180a2eb4daba2545f9b2d064668.png",[],{"type":18,"tag":26,"props":178,"children":179},{},[180],{"type":24,"value":181},"经过2个阶段，一个是验证阶段，一个是烧录阶段，不过，可惜的是最后一步出错了，百度一下有人说可以不用管，先尝试一下可不可以开机。",{"type":18,"tag":26,"props":183,"children":184},{},[185],{"type":18,"tag":131,"props":186,"children":188},{"alt":7,"src":187},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/01/17/6ebd37acd15c413b94ac4f14927d1bf1.png",[],{"type":18,"tag":26,"props":190,"children":191},{},[192],{"type":18,"tag":32,"props":193,"children":194},{},[195],{"type":24,"value":196},"机器组装",{"type":18,"tag":26,"props":198,"children":199},{},[200],{"type":24,"value":201},"插入USB的鼠标和键盘、显示器HDMI线、自带的type-c电源线后，即可以去插上电源了，一插入就自动启动了，没有什么开机键，即一通电就自动启动，但是有一个问题，等了5分钟除了风扇一直响外，显示器一直不显示，猜测是之前的烧录有问题。",{"type":18,"tag":26,"props":203,"children":204},{},[205],{"type":24,"value":206},"排查原因：",{"type":18,"tag":26,"props":208,"children":209},{},[210],{"type":18,"tag":131,"props":211,"children":213},{"alt":7,"src":212},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/01/17/6c2662bc57774ccfaab28585cc11d345.png",[],{"type":18,"tag":26,"props":215,"children":216},{},[217],{"type":18,"tag":131,"props":218,"children":220},{"alt":7,"src":219},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/01/17/53a269601e0d4c948622d17f5a7eff4e.png",[],{"type":18,"tag":26,"props":222,"children":223},{},[224],{"type":18,"tag":131,"props":225,"children":227},{"alt":7,"src":226},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/01/17/c3faced4b0c540c5b0ba4cc9dd83e089.png",[],{"type":18,"tag":229,"props":230,"children":231},"blockquote",{},[232],{"type":18,"tag":26,"props":233,"children":234},{},[235],{"type":24,"value":236},"因为手里没有备用的TF卡，只能通过第二种方案换一种工具来进行烧录试试，如果网友可以成功启动，就不需要下一步操作。",{"type":18,"tag":26,"props":238,"children":239},{},[240],{"type":18,"tag":32,"props":241,"children":242},{},[243],{"type":24,"value":244},"替换烧录工具",{"type":18,"tag":26,"props":246,"children":247},{},[248],{"type":24,"value":249},"在网友的推荐下，使用了另外一个烧录工具ascend，不过这个工具需要下一步下一步安装，就不过多复述了。",{"type":18,"tag":26,"props":251,"children":252},{},[253],{"type":18,"tag":131,"props":254,"children":256},{"alt":7,"src":255},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/01/17/ec68cdf697824b9b911f9896ddcde082.png",[],{"type":18,"tag":229,"props":258,"children":259},{},[260,265,270,275],{"type":18,"tag":26,"props":261,"children":262},{},[263],{"type":24,"value":264},"小细节：",{"type":18,"tag":26,"props":266,"children":267},{},[268],{"type":24,"value":269},"1、烧录时，提示U盘找不到，需要重新插入，此时，不用管他的提示。",{"type":18,"tag":26,"props":271,"children":272},{},[273],{"type":24,"value":274},"2、在烧录完成后，提示要格式化，也不需要管，因为在烧录时，就已经开始写成linux的启动镜像盘了，此时，在windows系统是无法识别的，是正常的。",{"type":18,"tag":26,"props":276,"children":277},{},[278],{"type":24,"value":279},"3、第一次烧的时候，看到东西就点，点快了，结果重新烧录了一次，需要注意一下。",{"type":18,"tag":26,"props":281,"children":282},{},[283],{"type":24,"value":284},"通过重新烧录之后，我们的系统就可以正常启动了，启动时，需要输入一个用户名密码，非常的开心，后面使用SSH也需要使用到。",{"type":18,"tag":77,"props":286,"children":287},{},[288,293],{"type":18,"tag":81,"props":289,"children":290},{},[291],{"type":24,"value":292},"用户名：HwHiAiUser",{"type":18,"tag":81,"props":294,"children":295},{},[296],{"type":24,"value":297},"密码：Mind@123",{"type":18,"tag":26,"props":299,"children":300},{},[301],{"type":18,"tag":131,"props":302,"children":304},{"alt":7,"src":303},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/01/17/c32fbeb1acac4eb080610c2132c2adbd.png",[],{"type":18,"tag":26,"props":306,"children":307},{},[308],{"type":18,"tag":32,"props":309,"children":310},{},[311],{"type":24,"value":312},"链接Wifi，使用SSH登录",{"type":18,"tag":26,"props":314,"children":315},{},[316],{"type":24,"value":317},"因为没有联网，所以，只能在显示器上进行Wifi的联网操作，联网后输入ifconfig查看Wifi的IP地址，在同一个局域网，使用ping ip地址保证网是能连通的情况下，我们就可以使用我工作的主机来SSH登录到开发板子上了，这样操作起来比较便捷。",{"type":18,"tag":319,"props":320,"children":322},"pre",{"code":321},"# 查看主机IP地址\nifconfig\n\n# 检查网络是否连通\nping 192.xx.xx.xx\n",[323],{"type":18,"tag":324,"props":325,"children":326},"code",{"__ignoreMap":7},[327],{"type":24,"value":321},{"type":18,"tag":26,"props":329,"children":330},{},[331],{"type":24,"value":332},"使用ssh远程链接到OrangePi AIpro开发板上，接下来可以进行环境的初始化，安装一些相关的依赖环境。",{"type":18,"tag":26,"props":334,"children":335},{},[336],{"type":18,"tag":131,"props":337,"children":339},{"alt":7,"src":338},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/01/17/1ebf0318bed042bd915287217d3111ee.png",[],{"type":18,"tag":229,"props":341,"children":342},{},[343],{"type":18,"tag":26,"props":344,"children":345},{},[346],{"type":24,"value":347},"硬件要求：产品硬件平台为Ascend，因为样例使用到opencv，ffmpeg等第三方依赖进行图像等处理，所以需要在运行之前，根据本文指导安装第三方依赖并进行环境配置。由于python的开发过程不依赖于编译，所以第三方依赖安装都在运行环境进行操作，第三方依赖python-acllite 请参考第三方依赖选择需要的依赖完成安装。",{"type":18,"tag":26,"props":349,"children":350},{},[351],{"type":24,"value":352},"配置编译依赖的头文件与库文件路径：",{"type":18,"tag":319,"props":354,"children":356},{"code":355},"export DDK_PATH=/usr/local/Ascend/ascend-toolkit/latest\nexport NPU_HOST_LIB=$DDK_PATH/runtime/lib64/stub\n",[357],{"type":18,"tag":324,"props":358,"children":359},{"__ignoreMap":7},[360],{"type":24,"value":355},{"type":18,"tag":26,"props":362,"children":363},{},[364],{"type":24,"value":365},"运行环境执行以下命令安装相关依赖及ffmpeg：",{"type":18,"tag":319,"props":367,"children":369},{"code":368},"sudo apt-get install ffmpeg libavcodec-dev libswscale-dev libavdevice-dev\n",[370],{"type":18,"tag":324,"props":371,"children":372},{"__ignoreMap":7},[373],{"type":24,"value":368},{"type":18,"tag":26,"props":375,"children":376},{},[377],{"type":24,"value":378},"源码安装ACLLite库，获取ACLLite 源码，python acllite库以源码方式提供，安装时将acllite目录拷贝到运行环境的第三方库目录：",{"type":18,"tag":319,"props":380,"children":382},{"code":381},"git clone https://gitee.com/ascend/ACLLite.git\n# 进入到项目目录进行编译, 会编译到上面设置的lib 目录\nbash build_so.sh\n",[383],{"type":18,"tag":324,"props":384,"children":385},{"__ignoreMap":7},[386],{"type":24,"value":381},{"type":18,"tag":26,"props":388,"children":389},{},[390],{"type":24,"value":391},"现在基本的初始化环境已经配置完成了，接下来就是拉取代码的准备和编译了。",{"type":18,"tag":26,"props":393,"children":394},{},[395],{"type":18,"tag":32,"props":396,"children":397},{},[398],{"type":24,"value":399},"模型转换",{"type":18,"tag":229,"props":401,"children":402},{},[403],{"type":18,"tag":26,"props":404,"children":405},{},[406],{"type":24,"value":407},"注意：如果是昇腾软件栈的AI推理，需要把原始网络模型 (可能是 PyTorch 的、TensorFlow，也有可能是Caffe的等) 转换成 .om 模型，才能调用昇腾的aclmdlExecute 等模型执行接口在进行模型推理。",{"type":18,"tag":26,"props":409,"children":410},{},[411],{"type":24,"value":412},"这个模型转换的过程就要用到 ATC 工具，目前 ATC 工具直接支持从 Caffe、ONNX、TensorFlow 以及 MindSpore模型的转换，所以如果的训练框架是 PyTorch，则需要做 torch.onnx.export 操作导出成ONNX模型后才能使用ATC工具。",{"type":18,"tag":26,"props":414,"children":415},{},[416],{"type":24,"value":417},"什么是ATC：",{"type":18,"tag":26,"props":419,"children":420},{},[421],{"type":24,"value":422},"昇腾张量编译器（Ascend Tensor Compiler，简称ATC）是昇腾模型转换工具，它可以将开源框架的网络模型（例如TensorFlow、ONNX等）转换为昇腾AI处理器支持的模型文件（.om格式），用于后续的模型推理。",{"type":18,"tag":26,"props":424,"children":425},{},[426],{"type":18,"tag":131,"props":427,"children":429},{"alt":7,"src":428},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/01/17/8d7d93cc8ff44b16b792fb140810f82e.png",[],{"type":18,"tag":26,"props":431,"children":432},{},[433],{"type":24,"value":434},"模型转换过程中，ATC会进行算子调度优化、权重数据重排、内存使用优化等操作，对开源框架的网络模型做进一步调优，使其高效地在昇腾AI处理器上执行。",{"type":18,"tag":26,"props":436,"children":437},{},[438],{"type":24,"value":439},"简单的来讲，就是可以将开源框架的网络模型（例如TensorFlow、ONNX等）转换为昇腾AI处理器支持的模型文件（.om格式），用于后续的模型推理，而Orange Ai pro 需要的是.om的格式，所以，需要把pt 的模型转成onnx 格式，再利用平台的 Atc工具转成.om的格式。",{"type":18,"tag":319,"props":441,"children":443},{"code":442},"# 为了方便下载，在这里直接给出原始模型下载及模型转换命令,可以直接拷贝执行\ncd model    \nwget https://obs-9be7.obs.cn-east-2.myhuaweicloud.com:443/003_Atc_Models/AE/ATC%20Model/garbage/mobilenetv2.air  \nwget https://obs-9be7.obs.cn-east-2.myhuaweicloud.com/models/garbage_picture/insert_op_yuv.cfg\natc --model=./mobilenetv2.air --framework=1 --output=garbage_yuv --soc_version= Ascend310B4 --insert_op_conf=./insert_op_yuv.cfg --input_shape=\"data:1,3,224,224\" --input_format=NCHW\n",[444],{"type":18,"tag":324,"props":445,"children":446},{"__ignoreMap":7},[447],{"type":24,"value":442},{"type":18,"tag":26,"props":449,"children":450},{},[451],{"type":24,"value":452},"–framework：原始网络模型框架类型，0：Caffe，1：MindSpore，3：TensorFlow，5：ONNX –soc_version：模型转换时昇腾AI处理器的版本，例如Ascend310B4 –model：原始网络模型文件，含扩展名 –output：转换后的*.om模型文件路径，含文件名，转换成功后，模型文件名自动以.om后缀结尾 –insert_op_conf: 模型相关的配置文件，包含图像大小，预处理等参数。",{"type":18,"tag":26,"props":454,"children":455},{},[456],{"type":18,"tag":131,"props":457,"children":459},{"alt":7,"src":458},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/01/17/bc26e146f6704f3abfa3586ac340f4ab.png",[],{"type":18,"tag":26,"props":461,"children":462},{},[463],{"type":24,"value":464},"mobilenetv2.air文件是一个模型文件：",{"type":18,"tag":26,"props":466,"children":467},{},[468],{"type":24,"value":469},"‌①通常是在模型转换过程中生成的。‌",{"type":18,"tag":26,"props":471,"children":472},{},[473],{"type":24,"value":474},"②这个文件包含了模型的架构和权重信息，‌可以用于在特定硬件或软件平台上进行推理。",{"type":18,"tag":26,"props":476,"children":477},{},[478],{"type":24,"value":479},"‌③将MobileNetV2模型从PyTorch格式转换为ONNX格式，‌然后再进一步转换为适用于Ascend310芯片的模型格式，‌最终生成了mobilenetv2.air文件。‌",{"type":18,"tag":26,"props":481,"children":482},{},[483],{"type":24,"value":484},"insert_op_yuv.cfg文件则是一个配置文件：",{"type":18,"tag":26,"props":486,"children":487},{},[488],{"type":24,"value":489},"①用于定义在模型转换或推理过程中需要插入的特定操作或配置。",{"type":18,"tag":26,"props":491,"children":492},{},[493],{"type":24,"value":494},"‌②这个文件可能包含了针对特定硬件或软件平台的优化设置，‌以确保模型能够在特定环境下高效运行。‌",{"type":18,"tag":26,"props":496,"children":497},{},[498],{"type":24,"value":499},"③可能包含了关于输入数据的预处理、‌输出数据的后处理以及可能的硬件加速指令等信息。",{"type":18,"tag":26,"props":501,"children":502},{},[503],{"type":24,"value":504},"④被用于Ascend芯片上的模型转换过程中，‌以实现模型的优化和加速。‌",{"type":18,"tag":26,"props":506,"children":507},{},[508],{"type":24,"value":509},"总结，通过使用Ascend的实例转换模型非常的快，只用了2分钟就完成了操作，而用香橙派OrangePi AIpro，由于硬件配置原因，转了半小时也没反应，果断换了实例。",{"type":18,"tag":26,"props":511,"children":512},{},[513],{"type":18,"tag":32,"props":514,"children":515},{},[516],{"type":24,"value":517},"官方的Ai Demo项目",{"type":18,"tag":319,"props":519,"children":521},{"code":520},"# 开发环境，非root用户命令行中执行以下命令下载源码仓。   \ncd ${HOME}    \ngit clone https://gitee.com/ascend/samples.git\n# 将ZIP包上传到开发环境中的目录中，【例如：${HOME}/ascend-samples-master.zip】\nunzip ascend-samples-master.zip\n# 进入项目\ncd python/contrib/garbage_picture\n",[522],{"type":18,"tag":324,"props":523,"children":524},{"__ignoreMap":7},[525],{"type":24,"value":520},{"type":18,"tag":26,"props":527,"children":528},{},[529],{"type":24,"value":530},"在data中添加一些需要识别的图片：",{"type":18,"tag":26,"props":532,"children":533},{},[534],{"type":18,"tag":131,"props":535,"children":537},{"alt":7,"src":536},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/01/17/fe84ff95abd546bb95043c1d9fc56134.png",[],{"type":18,"tag":26,"props":539,"children":540},{},[541],{"type":24,"value":542},"运行可执行文件：",{"type":18,"tag":319,"props":544,"children":546},{"code":545},"python3 classify_test.py ../data/\n",[547],{"type":18,"tag":324,"props":548,"children":549},{"__ignoreMap":7},[550],{"type":24,"value":545},{"type":18,"tag":26,"props":552,"children":553},{},[554],{"type":18,"tag":131,"props":555,"children":557},{"alt":7,"src":556},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/01/17/1a8f9188f03e40bdaf355650c31e14df.png",[],{"type":18,"tag":26,"props":559,"children":560},{},[561],{"type":24,"value":562},"提示：这里如果使用英文的字体是写不出来中文的，可以确认系统是否已经安装了中文字体，可以使用 fc-list 命令来查看当前系统上安装的所有字体，如下所示：",{"type":18,"tag":319,"props":564,"children":566},{"code":565},"fc-list\n",[567],{"type":18,"tag":324,"props":568,"children":569},{"__ignoreMap":7},[570],{"type":24,"value":565},{"type":18,"tag":26,"props":572,"children":573},{},[574],{"type":18,"tag":32,"props":575,"children":576},{},[577],{"type":24,"value":578},"查看结果",{"type":18,"tag":26,"props":580,"children":581},{},[582],{"type":24,"value":583},"运行完成后，会在out目录下生成带推理结果的jpg图片。",{"type":18,"tag":26,"props":585,"children":586},{},[587],{"type":18,"tag":131,"props":588,"children":590},{"alt":7,"src":589},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2025/01/17/4202ae287e8f4c1696f510d5750e5330.png",[],{"type":18,"tag":26,"props":592,"children":593},{},[594],{"type":24,"value":595},"总结，通过上面的实验，可以看到我们可以很好的在香橙派OrangePi AIpro上运行基于MobileNetV2垃圾分类项目，可以看到有一个电池的图片没有识别正确，其他都正确，识别的结果为“牙刷”，可以增加多一点的图片，再自己训练一下。",{"title":7,"searchDepth":597,"depth":597,"links":598},4,[],"markdown","content:technology-blogs:zh:3580.md","content","technology-blogs/zh/3580.md","technology-blogs/zh/3580","md",1776506131498]