# 使用 Neural Processing SDK 进行图像分类和编码

Source: [https://docs.qualcomm.com/doc/80-70014-50Y/topic/single-camera-stream-with-image-classification-and-encode-with-mobilenet-v1.html](https://docs.qualcomm.com/doc/80-70014-50Y/topic/single-camera-stream-with-image-classification-and-encode-with-mobilenet-v1.html)

这些用例使用 Inceptionv3 图像分类模型和 Qualcomm Neural Processing SDK，将场景分类为单个摄像头流，并叠加或合成分类标签，接着对流进行编码。

可以使用 TensorFlow 获取任何公开可用的分类模型，并将其转换为 `.dlc` 的格式，如 [TensorFlow 模型转换](https://docs.qualcomm.com/bundle/publicresource/topics/80-63442-2/model_conv_tensorflow.html)中所述。

## 变体 1：使用 qtioverlay 插件来应用分类叠加

使用以下命令来执行用例：

    setprop persist.overlay.use_c2d_blit 2Copy to clipboard

    gst-launch-1.0 -e \
    qtiqmmfsrc name=camsrc ! video/x-raw\(memory:GBM\),format=NV12,width=1280,height=720,framerate=30/1,compression=ubwc ! queue ! tee name=split \
    split. ! queue ! qtimetamux name=metamux ! queue ! qtioverlay ! queue ! video/x-raw\(memory:GBM\),format=NV12,width=1280,height=720,interlace-mode=progressive,colorimetry=bt601 ! v4l2h264enc capture-io-mode=5 output-io-mode=5 ! h264parse ! queue ! mp4mux ! queue ! filesink location=/opt/video.mp4 \
    split. ! queue ! qtimlvconverter ! queue ! qtimlsnpe delegate=dsp model=/opt/inceptionv3.dlc ! queue ! qtimlvclassification threshold=40.0 results=2 module=mobilenet labels=/opt/classification.labels ! text/x-raw ! queue ! metamux.Copy to clipboard

如需停止用例，请按下 CTRL + C。

Figure : 分类叠加和编码的 pipeline
                ![](data:image/png;base64,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)

下图显示了用例执行流程：

1. 对从摄像头源传来的视频流中的场景进行分类。
2. 使用 overlaylib 叠加分类标签。
3. 将此数据流编码为 H.264 码流。
4. 将数据流多路复用到 MP4 容器中，并将其存储为 MP4 文件。

pipeline 执行的顺序处理阶段如下表所示：

| 处理过程 | 说明 |
| --- | --- |
| Source | <ol class="ol" id="single-camera-stream-with-image-classification-and-encode-with-mobilenet-v1__ol_g41_cf5_vbc"><br>                                    <li class="li">从摄像头源插件采集视频流，并创建两个副本：<ul class="ul" id="single-camera-stream-with-image-classification-and-encode-with-mobilenet-v1__ol_kh3_dyv_r1c"><br>                                            <li class="li">一个视频流被发送到 qtimetamux 插件以保留视频流。</li><br><br>                                            <li class="li">另一个视频流被发送到 ML 推理 pipeline。</li><br><br>                                        </ul><br></li><br><br>                                </ol> |
| **预处理** | **预处理** |
| [qtimlvconverter](https://docs.qualcomm.com/doc/80-70014-50Y/topic/qtimlvconverter.html) | <ol class="ol" id="single-camera-stream-with-image-classification-and-encode-with-mobilenet-v1__ol_xsf_q5l_vbc"><br>                                    <li class="li">在其接收设备接插口上接收视频流。</li><br><br>                                    <li class="li">执行预处理：<ul class="ul" id="single-camera-stream-with-image-classification-and-encode-with-mobilenet-v1__ul_ff2_twl_vbc"><br>                                            <li class="li">色彩转换</li><br><br>                                            <li class="li">缩小/放大</li><br><br>                                            <li class="li">在模型需要浮点值作为输入时，对流数据进行归一化</li><br><br>                                        </ul><br></li><br><br>                                    <li class="li">在其源接插口上将视频流转换为张量数据。<p class="p">分类模型使用此张量数据进行推理。</p><br></li><br><br>                                </ol> |
| **推理** | **推理** |
| [qtimlsnpe](https://docs.qualcomm.com/doc/80-70014-50Y/topic/qtimlsnpe.html) | <ol class="ol" id="single-camera-stream-with-image-classification-and-encode-with-mobilenet-v1__ol_bwn_s5l_vbc"><br>                                    <li class="li">加载模型。</li><br><br>                                    <li class="li">为选择的 delegate 修改图形。</li><br><br>                                    <li class="li">在其接收设备接插口上接收张量数据。</li><br><br>                                    <li class="li">执行推理并在其源接插口上生成包含推理结果的张量数据。</li><br><br>                                </ol> |
| **后处理** | **后处理** |
| [qtimlvclassification](https://docs.qualcomm.com/doc/80-70014-50Y/topic/qtimlvclassification.html) | <ol class="ol" id="single-camera-stream-with-image-classification-and-encode-with-mobilenet-v1__ol_gr1_w5l_vbc"><br>                                    <li class="li">从其接收设备接插口上的分类模型接收推理张量。</li><br><br>                                    <li class="li">将张量转换为视频或文本等格式，稍后可由多媒体插件进行处理。</li><br><br>                                    <li class="li">将阈值应用于所选的结果数。</li><br><br>                                    <li class="li">加载分类模型的相应模块。<p class="p">在此用例中，qtimlvclassification 执行以下操作：</p><ol class="ol" type="a" id="single-camera-stream-with-image-classification-and-encode-with-mobilenet-v1__ol_rrb_1xl_vbc"><br>                                            <li class="li">加载模型的子模块。</li><br><br>                                            <li class="li">将结果生成为带有分类标签的视频帧。</li><br><br>                                            <li class="li">接着发送到 qtimetamux 的接收设备接插口。</li><br><br>                                        </ol><br></li><br><br>                                </ol> |
| [qtimetamux](https://docs.qualcomm.com/doc/80-70014-50Y/topic/qtimetamux.html) | <ol class="ol" id="single-camera-stream-with-image-classification-and-encode-with-mobilenet-v1__ol_ll3_x5l_vbc"><br>                                    <li class="li">在接收设备接插口上接收视频流和文本流，以及与视频流相对应的分类结果。</li><br><br>                                    <li class="li">在其接收设备接插口上生成包含视频流内容的 GST 缓冲区。</li><br><br>                                    <li class="li">将数据接收设备接插口上的分类结果添加到其源接插口上的 GST 缓冲区元数据（元数据多路复用）。</li><br><br>                                </ol> |
| [qtioverlay](https://docs.qualcomm.com/doc/80-70014-50Y/topic/qtioverlay.html) | <ol class="ol" id="single-camera-stream-with-image-classification-and-encode-with-mobilenet-v1__ol_wst_y5l_vbc"><br>                                    <li class="li">接收多路复用流。</li><br><br>                                    <li class="li">使用 CL 将分类标签叠加在 VideoFrame 上。</li><br><br>                                    <li class="li">在其源接插口上生成带有叠加层的 GST 缓冲区。</li><br><br>                                </ol> |
| [v4l2h264enc](https://docs.qualcomm.com/doc/80-70014-50Y/topic/v4l2h264enc.html) | <ol class="ol" id="single-camera-stream-with-image-classification-and-encode-with-mobilenet-v1__ol_h41_cf5_vbc"><br>                                    <li class="li">将参数应用于在接收设备接插口上接收的视频流的每一帧。</li><br><br>                                    <li class="li">将其编码为比特流，并通过其源接插口发送。</li><br><br>                                </ol> |
| h264parse | 将与比特流对应的其他信息添加到 GStreamer 缓冲区元数据。 |
| mp4mux | 接收这些缓冲区并创建具有格式规范缓冲区的容器。 |
| **输出** | **输出** |
| Filesink | 将生成的数据流存储在 /opt/video.mp4 文件中。 |
| Playback | 使用以下命令从主机拉取 video.mp4 并在媒体播放器应用上播放：<br>`scp root@<IP address of target device>:/opt/ <destination directory>` |

## 变体 2：使用 qtivcomposer 混合原始帧与分类掩码

使用以下命令来执行用例：

    gst-launch-1.0 -e --gst-debug=2 \
    qtiqmmfsrc name=camsrc ! video/x-raw\(memory:GBM\),format=NV12,width=1280,height=720,framerate=30/1,compression=ubwc ! queue ! tee name=split \
    split. ! queue ! qtivcomposer name=mixer sink_1::position="<30, 30>" sink_1::dimensions="<320, 180>" ! queue ! video/x-raw\(memory:GBM\),format=NV12,width=1920,height=1080,interlace-mode=progressive,colorimetry=bt601 ! v4l2h264enc capture-io-mode=5 output-io-mode=5 ! h264parse ! queue ! mp4mux ! queue ! filesink location=/opt/video.mp4 \
    split. ! queue ! qtimlvconverter ! queue ! qtimlsnpe delegate=dsp model=/opt/inceptionv3.dlc ! queue ! qtimlvclassification threshold=40.0 results=2 module=mobilenet labels=/opt/classification.labels ! video/x-raw,format=BGRA,width=640,height=360 ! queue ! mixer.Copy to clipboard

如需停止用例，请按下 CTRL + C。

Figure : 使用 qtivcomposer 进行分类和编码的 pipeline
                ![](data:image/png;base64,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)

下图显示了用例执行流程：
- 对从摄像头源传来的视频流中的场景进行分类。
- 使用 qtivcomposer 合成分类标签和视频流。
- 将此数据流编码为 H.264 比特流。
- 将数据流多路复用到 MP4 容器中，并将其存储为 MP4 文件。

pipeline 执行的顺序处理阶段如下表所示：

| 处理过程 | 说明 |
| --- | --- |
| Source | <ol class="ol" id="single-camera-stream-with-image-classification-and-encode-with-mobilenet-v1__ol_acs_jk5_vbc"><br>                                    <li class="li">从摄像头源插件采集视频流，并创建两个副本：<ul class="ul" id="single-camera-stream-with-image-classification-and-encode-with-mobilenet-v1__ul_bcs_jk5_vbc"><br>                                            <li class="li">一个视频流被发送到 qtivcomposer 插件以保留视频流。</li><br><br>                                            <li class="li">另一个视频流被发送到 ML 推理 pipeline。</li><br><br>                                        </ul><br></li><br><br>                                </ol> |
| **预处理** | **预处理** |
| [qtimlvconverter](https://docs.qualcomm.com/doc/80-70014-50Y/topic/qtimlvconverter.html) | <ol class="ol" id="single-camera-stream-with-image-classification-and-encode-with-mobilenet-v1__ol_ccs_jk5_vbc"><br>                                    <li class="li">在其接收设备接插口上接收视频流。</li><br><br>                                    <li class="li">执行预处理：<ul class="ul" id="single-camera-stream-with-image-classification-and-encode-with-mobilenet-v1__ul_dcs_jk5_vbc"><br>                                            <li class="li">色彩转换</li><br><br>                                            <li class="li">缩小/放大</li><br><br>                                            <li class="li">在模型需要浮点值作为输入时，对流数据进行归一化</li><br><br>                                        </ul><br></li><br><br>                                    <li class="li">在其源接插口上将视频流转换为张量数据。<p class="p">分类模型使用此张量数据进行推理。</p><br></li><br><br>                                </ol> |
| **推理** | **推理** |
| [qtimlsnpe](https://docs.qualcomm.com/doc/80-70014-50Y/topic/qtimlsnpe.html) | <ol class="ol" id="single-camera-stream-with-image-classification-and-encode-with-mobilenet-v1__ol_ecs_jk5_vbc"><br>                                    <li class="li">加载模型。</li><br><br>                                    <li class="li">为选择的 delegate 修改图形。</li><br><br>                                    <li class="li">在其接收设备接插口上接收张量数据。</li><br><br>                                    <li class="li">执行推理并在其源接插口上生成包含推理结果的张量数据。</li><br><br>                                </ol> |
| **后处理** | **后处理** |
| [qtimlvclassification](https://docs.qualcomm.com/doc/80-70014-50Y/topic/qtimlvclassification.html) | <ol class="ol" id="single-camera-stream-with-image-classification-and-encode-with-mobilenet-v1__ol_fcs_jk5_vbc"><br>                                    <li class="li">从其接收设备接插口上的分类模型接收推理张量。</li><br><br>                                    <li class="li">将张量转换为视频或文本等格式，稍后可由多媒体插件进行处理。</li><br><br>                                    <li class="li">将阈值应用于所选的结果数。</li><br><br>                                    <li class="li">加载分类模型的相应模块。<p class="p">在此用例中，qtimlvclassification 执行以下操作：</p><ol class="ol" type="a" id="single-camera-stream-with-image-classification-and-encode-with-mobilenet-v1__ol_gcs_jk5_vbc"><br>                                            <li class="li">加载模型的子模块。</li><br><br>                                            <li class="li">将结果生成为带有分类标签的视频帧。</li><br><br>                                            <li class="li">接着发送到 qtivcomposer 的接收设备接插口。</li><br><br>                                        </ol><br></li><br><br>                                </ol> |
| [qtivcomposer](https://docs.qualcomm.com/doc/80-70014-50Y/topic/qtivcomposer.html) | <ol class="ol" id="single-camera-stream-with-image-classification-and-encode-with-mobilenet-v1__ol_gv1_wjk_5bc"><br>                                    <li class="li">在其接收设备接插口上接收原始视频流和带有分类结果的视频流。</li><br><br>                                    <li class="li">在其源接插口上生成 GST 缓冲区，其内容由来自其接收设备接插口的视频流组成。</li><br><br>                                </ol> |
| [v4l2h264enc](https://docs.qualcomm.com/doc/80-70014-50Y/topic/v4l2h264enc.html) | <ol class="ol" id="single-camera-stream-with-image-classification-and-encode-with-mobilenet-v1__ol_jcs_jk5_vbc"><br>                                    <li class="li">将参数应用于在接收设备接插口上接收的视频流的每一帧。</li><br><br>                                    <li class="li">将其编码为比特流，并通过其源接插口发送。</li><br><br>                                </ol> |
| h264parse | 将与比特流对应的其他信息添加到 GStreamer 缓冲区元数据。 |
| mp4mux | 接收这些缓冲区并创建具有格式规范缓冲区的容器。 |
| **输出** | **输出** |
| Filesink | 将生成的数据流存储在 /opt/video.mp4 文件中。 |
| Playback | 使用以下命令从主机拉取 video.mp4 并在媒体播放器应用上播放：<br>`scp root@<IP address of target device>:/opt/ <destination directory>` |

**Parent Topic:** [Qualcomm Neural Processing SDK 用例](https://docs.qualcomm.com/doc/80-70014-50Y/topic/qualcomm-neural-processing-sdk-use-cases.html)

Last Published: Nov 11, 2025

[Previous Topic
使用 Neural Processing SDK 进行图像分类和显示](https://docs.qualcomm.com/bundle/publicresource/80-70014-50Y/topics/single-camera-stream-with-image-classification-and-display-with-mobilenet-v1.md) [Next Topic
使用 Neural Processing SDK 进行目标检测和显示](https://docs.qualcomm.com/bundle/publicresource/80-70014-50Y/topics/single-camera-stream-with-object-detection-and-display-with-mobilenet-v2-ssd.md)