Changes

In [[SBCX-TN-005: Using TensorFlow to implement a Deep Learning image classifier based on Azure Custom Vision-generated model|this Technical Note (SBCX-TN-005)]] (TN for short), a simple image classifier was implemented on the [[:Category:AxelLite|Axel Lite SoM]]. In [[MISC-TN-010: Using NXP eIQ Machine Learning Development Environment with Mito8M SoM|this other]], it is illustrated how to run [https://www.nxp.com/design/software/development-software/eiq-ml-development-environment:EIQ NXP eIQ Machine Learning software] on i.MX8M-powered [[:Category:Mito8M|Mito8M SoM]]. This article combines the results shown in the TN's just mentioned. In other words, it describes how to run the same image classifier used in SBCX-TN-005 with the eIQ software stack. The outcome is an optimized C++ imaging classification application running on Mito8M SoM, which makes use of the eIQ software stack. In terms of hardware and software, the testbed used is the same described [[MISC-TN-010: Using NXP eIQ Machine Learning Development Environment with Mito8M SoM|here]]. ==Workflow and resulting block diagram==The following picture shows the block diagram of the resulting application and part of the workflow used to build it.  [[File:MISC-TN-011-image-classifier.png|thumb|center|600px|Block diagram of the image classifier]]  First of all, the TensorFlow (TF) model generated with Microsoft Azure Custom Vision was converted into the TensorFlow Lite (TFL) format. Then, a new C++ application was written, using the examples provided by TFL as starting points. After debugging this application on a host PC, it was migrated to the edge device (a Mito8M-powered platform, in our case) where it was natively built. The root file system for eIQ, in fact, provides the native C++ compiler as well. For the sake of completeness, the following images show the graphs of the original TF model and the converted TFL model (click to enlarge). {| class="wikitable" style="margin: auto;"|+!Original TF model's graph!Converted TFL model's graph|-|[[File:Image-classifier-azure-model.pb.png|thumb|center|150px]]|[[File:Image-classifier-azure-converted model.tflite.png|thumb|center|370px]]|}

In this [[MISC==Running the application==The following block shows the execution of the classifier on the embedded platform:<pre class="board-TNterminal">root@mito8m:~/devel/image_classifier_eIQ# ./image_classifier_cv converted_model.tflite labels.txt testdata/red-010apple1.jpgOriginal image size: 600x600x3Cropped image size: 600x600x3Resized image size: 224x224x3Input tensor index: 0Input tensor name: PlaceholderFilling time: 25.3169 msInference time: 276.121 msTotal prediction time: Using NXP eIQ Machine Learning Development Environment with Mito8M SoM|TN (MISC-TN-010)]], it is illustrated how to run NXP eIQ Machine Learning software on i301.MX8M-powered [[438 msOutput tensor index: 406Output tensor name:Categorymodel_outputsTop results:Mito8M|Mito8M SoM]] 0.997172 Red Apple 0.00214239 Green Apple</pre>

This article combines the results shown in the TN's just mentioned. In other words, it describes how The prediction time is cut by about 88% compared to run the same image classifier used in [[SBCX-TN-005 with the eIQ software stack: Using TensorFlow to implement a Deep Learning image classifier based on Azure Custom Vision-generated model|this implementation]]. Of course, this is due to several factors. The outcome most relevant ones are:* i.MX8M is an optimized imaging classification faster than i.MX6Q* The application is written in C++ running on Mito8M SoM and that makes use of not in Python* The TF model was replaced with a TFL model, which is inherently more suited for ARM-based devices* The middleware provided by NXP eIQ software stackis optimized for their SoC's.