Edge Impulse Experts / AI-based Aquatic Air Bubble Detection with Ultrasonic Imaging Public

An impulse takes raw data, uses signal processing to extract features, and then uses a learning block to classify new data.

Description	Recommended
Pre-processed features Officially Supported Processes pre-processed features, or non time-series data.
Time series data Officially Supported Operates on time series sensor data like vibration or audio data. Lets you slice up data into windows.
Images Officially Supported Processes discrete images for object detection or classification.

Did you know? You can bring your own DSP code.

Description	Author	Recommended
Raw Data Officially Supported Use data without pre-processing. Useful if you want to use deep learning to learn features.	Edge Impulse
Edge Detection (created by louis) Enterprise Edge Detection for images	Edge Impulse Experts
PoseNet (uploaded by peter ing) Enterprise Pose estimation for images	Edge Impulse Experts
pose-akida Enterprise Pose estimation for images	Edge Impulse Experts
Raw Data Officially Supported Use data without pre-processing. Useful if you want to use deep learning to learn features.	Edge Impulse
Flatten Officially Supported Flatten an axis into a single value, useful for slow-moving averages like temperature data, in combination with other blocks.	Edge Impulse
Image Officially Supported Preprocess and normalize image data, and optionally reduce the color depth.	Edge Impulse
Audio (MFCC) Officially Supported Extracts features from audio signals using Mel Frequency Cepstral Coefficients, great for human voice.	Edge Impulse
Audio (MFE) Officially Supported Extracts a spectrogram from audio signals using Mel-filterbank energy features, great for non-voice audio.	Edge Impulse
Spectral Analysis Officially Supported Great for analyzing repetitive motion, such as data from accelerometers. Extracts the frequency and power characteristics of a signal over time.	Edge Impulse
Spectrogram Officially Supported Extracts a spectrogram from audio or sensor data, great for non-voice audio or data with continuous frequencies.	Edge Impulse
Audio (Syntiant) Officially Supported Syntiant only. Compute log Mel-filterbank energy features from an audio signal.	Syntiant
IMU (Syntiant) Officially Supported Syntiant only. Great for analyzing repetitive motion, such as data from accelerometers. Extracts the frequency and power characteristics of a signal over time.	Syntiant
Edge Detection (created by louis) Enterprise Edge Detection for images	Edge Impulse Experts
PoseNet (uploaded by peter ing) Enterprise Pose estimation for images	Edge Impulse Experts
pose-akida Enterprise Pose estimation for images	Edge Impulse Experts

Did you know? You can bring your own model in PyTorch, Keras or scikit-learn.

Description	Author	Recommended
Classification Officially Supported Learns patterns from data, and can apply these to new data. Great for categorizing movement or recognizing audio.	Edge Impulse
Regression Officially Supported Learns patterns from data, and can apply these to new data. Great for predicting numeric continuous values.	Edge Impulse
Anomaly Detection (GMM) Professional Enterprise Find outliers in new data. A Gaussian mixture model (GMM) models the shape of data using a probability distribution. New data that is unlikely according to this model can be considered anomalous.	Edge Impulse
Anomaly Detection (K-means) Officially Supported Find outliers in new data. Good for recognizing unknown states, and to complement classifiers. Works best with low dimensionality features like the output of the spectral features block.	Edge Impulse
Classification - BrainChip Akida™ Officially Supported Learns patterns from data, and can apply these to new data. Great for categorizing movement or recognizing audio. Only works with BrainChip AKD1000 MINI PCIe board.	BrainChip
Visual Anomaly Detection - FOMO-AD Enterprise Detect visual anomalies. Extracts visual features using a pre-trained backbone, and applies a scoring function to evaluate how anomalous a sample is by comparing the extracted features to the learned model. Does not require anomalous data.	Edge Impulse
Classification Officially Supported Learns patterns from data, and can apply these to new data. Great for categorizing movement or recognizing audio.	Edge Impulse
Transfer Learning (Images) Officially Supported Fine tune a pre-trained image classification model on your data. Good performance even with relatively small image datasets.	Edge Impulse
Object Detection (Images) Officially Supported Fine tune a pre-trained object detection model on your data. Good performance even with relatively small image datasets.	Edge Impulse
Regression Officially Supported Learns patterns from data, and can apply these to new data. Great for predicting numeric continuous values.	Edge Impulse
Transfer Learning (Keyword Spotting) Officially Supported Fine tune a pre-trained keyword spotting model on your data. Good performance even with relatively small keyword datasets.	Edge Impulse
Anomaly Detection (GMM) Professional Enterprise Find outliers in new data. A Gaussian mixture model (GMM) models the shape of data using a probability distribution. New data that is unlikely according to this model can be considered anomalous.	Edge Impulse
Anomaly Detection (K-means) Officially Supported Find outliers in new data. Good for recognizing unknown states, and to complement classifiers. Works best with low dimensionality features like the output of the spectral features block.	Edge Impulse
Classification - BrainChip Akida™ Officially Supported Learns patterns from data, and can apply these to new data. Great for categorizing movement or recognizing audio. Only works with BrainChip AKD1000 MINI PCIe board.	BrainChip
Transfer Learning (Images) - BrainChip Akida™ Officially Supported Fine tune a pre-trained image classification model on your data. Good performance even with relatively small image datasets. Only works with BrainChip AKD1000 MINI PCIe board.	BrainChip
Object Detection (Images) - BrainChip Akida™ Officially Supported Fine tune a pre-trained object detection model on your data. Good performance even with relatively small image datasets. Only works with BrainChip AKD1000 MINI PCIe board.	BrainChip
Visual Anomaly Detection - FOMO-AD Enterprise Detect visual anomalies. Extracts visual features using a pre-trained backbone, and applies a scoring function to evaluate how anomalous a sample is by comparing the extracted features to the learned model. Does not require anomalous data.	Edge Impulse
YOLOv5(added by Solomon Githu) Enterprise Object Detection using YOLOv5	Edge Impulse Experts
YOLOx Enterprise YOLOx fork for TI TDA4VM	Edge Impulse Experts

Time series data

Input axes (400)

p_0, p_1, p_2, p_3, p_4, p_5, p_6, p_7, p_8, p_9, p_10, p_11, p_12, p_13, p_14, p_15, p_16, p_17, p_18, p_19, p_20, p_21, p_22, p_23, p_24, p_25, p_26, p_27, p_28, p_29, p_30, p_31, p_32, p_33, p_34, p_35, p_36, p_37, p_38, p_39, p_40, p_41, p_42, p_43, p_44, p_45, p_46, p_47, p_48, p_49, p_50, p_51, p_52, p_53, p_54, p_55, p_56, p_57, p_58, p_59, p_60, p_61, p_62, p_63, p_64, p_65, p_66, p_67, p_68, p_69, p_70, p_71, p_72, p_73, p_74, p_75, p_76, p_77, p_78, p_79, p_80, p_81, p_82, p_83, p_84, p_85, p_86, p_87, p_88, p_89, p_90, p_91, p_92, p_93, p_94, p_95, p_96, p_97, p_98, p_99, p_100, p_101, p_102, p_103, p_104, p_105, p_106, p_107, p_108, p_109, p_110, p_111, p_112, p_113, p_114, p_115, p_116, p_117, p_118, p_119, p_120, p_121, p_122, p_123, p_124, p_125, p_126, p_127, p_128, p_129, p_130, p_131, p_132, p_133, p_134, p_135, p_136, p_137, p_138, p_139, p_140, p_141, p_142, p_143, p_144, p_145, p_146, p_147, p_148, p_149, p_150, p_151, p_152, p_153, p_154, p_155, p_156, p_157, p_158, p_159, p_160, p_161, p_162, p_163, p_164, p_165, p_166, p_167, p_168, p_169, p_170, p_171, p_172, p_173, p_174, p_175, p_176, p_177, p_178, p_179, p_180, p_181, p_182, p_183, p_184, p_185, p_186, p_187, p_188, p_189, p_190, p_191, p_192, p_193, p_194, p_195, p_196, p_197, p_198, p_199, p_200, p_201, p_202, p_203, p_204, p_205, p_206, p_207, p_208, p_209, p_210, p_211, p_212, p_213, p_214, p_215, p_216, p_217, p_218, p_219, p_220, p_221, p_222, p_223, p_224, p_225, p_226, p_227, p_228, p_229, p_230, p_231, p_232, p_233, p_234, p_235, p_236, p_237, p_238, p_239, p_240, p_241, p_242, p_243, p_244, p_245, p_246, p_247, p_248, p_249, p_250, p_251, p_252, p_253, p_254, p_255, p_256, p_257, p_258, p_259, p_260, p_261, p_262, p_263, p_264, p_265, p_266, p_267, p_268, p_269, p_270, p_271, p_272, p_273, p_274, p_275, p_276, p_277, p_278, p_279, p_280, p_281, p_282, p_283, p_284, p_285, p_286, p_287, p_288, p_289, p_290, p_291, p_292, p_293, p_294, p_295, p_296, p_297, p_298, p_299, p_300, p_301, p_302, p_303, p_304, p_305, p_306, p_307, p_308, p_309, p_310, p_311, p_312, p_313, p_314, p_315, p_316, p_317, p_318, p_319, p_320, p_321, p_322, p_323, p_324, p_325, p_326, p_327, p_328, p_329, p_330, p_331, p_332, p_333, p_334, p_335, p_336, p_337, p_338, p_339, p_340, p_341, p_342, p_343, p_344, p_345, p_346, p_347, p_348, p_349, p_350, p_351, p_352, p_353, p_354, p_355, p_356, p_357, p_358, p_359, p_360, p_361, p_362, p_363, p_364, p_365, p_366, p_367, p_368, p_369, p_370, p_371, p_372, p_373, p_374, p_375, p_376, p_377, p_378, p_379, p_380, p_381, p_382, p_383, p_384, p_385, p_386, p_387, p_388, p_389, p_390, p_391, p_392, p_393, p_394, p_395, p_396, p_397, p_398, p_399

Window size

Window increase

Frequency (Hz)

Invalid frequency

Zero-pad data

Classification

Name

Input features

Raw data

Output features

2 (bubble, normal)

Output features

2 (bubble, normal)

No data collected yet

You'll need some training data to design your first impulse.

Go to data acquisition

We're always looking for ways to improve Edge Impulse. If you have any feedback, please let us know!

Subject

This field is required

How was your experience?

This field is required

You'll need a free Edge Impulse account to clone this project. Creating an account lets you add your own data, modify models, and join a community of thousands of embedded machine learning developers!

Target device

Define your target device requirements to inform model optimizations and performance calculations. No device yet? Use the default settings which you can change at any time.

Target device

Processor family

Clock rate

| MHz

Max

Custom device name

Application budget

Specify the available RAM and ROM for the model's operation, along with the maximum allowed latency for your specific application. Not sure yet? Start with the defaults and modify them later on.

RAM

| KB

Max

ROM

| KB

Max

Latency

| ms

Max

YEARLY

$400 /month

billed annually

SAVE 15%

MONTHLY

$475 /month

billed monthly

Time series data

Raw Data

Classification

Output features

No data collected yet