deepakrana47 / wheat-quality-detector-2 Goto Github PK

View Code? Open in Web Editor NEW

18.0 3.0 3.0 89.42 MB

Wheat classification based on the provided dataset

License: MIT License

Python 100.00%

wheat wheat-classification wheat-quality-detection

wheat-quality-detector-2's Introduction

Wheat-quality-detector-2

Inspired by link

Description:

This wheat quality detection test we are using to identify the quality of given wheat grains image. The dataset we used is provided by link.

The wheat quality detection problem is divided into two sub problems given as following:

Two classs classification i.e. a healthy grain or other.
Five class slassification i.e. a healthy grain, damaged grain, foreign partical, broken grain and grain cover.

The dataset we used for training (is the single grain or other images) extracted for above mentioned dataset.

Requirement:

opencv-python
keras
tensorflow
matplotlib

Tested with python3.5

A Glance

For two class classification:

 $ python classifier_2_v2.py
 68/68 [==============================] - 0s 499us/step - accuracy: 0.9020 - loss: 0.2475
 MLP Test loss: 0.247524231672287
 MLP Test accuracy: 0.9019879698753357

For five class classification:

 $ python classifier_5_v2.py
 65/65 [==============================] - 0s 532us/step - loss: 0.4837 - accuracy: 0.8254
 MLP Test loss: 0.483661413192749
 MLP Test accuracy: 0.8253890872001648

For performing a saimple test:

 $ python cmd_wheat_quality_detector_v2.py
 Enter the file(wheat image) location to dectect : test_2.jpg
 Segmentation in process...
 Level 1 segmentation Finished:
 Rejected segment: 1
 Level 2 segmentation Finished:
 Rejected segment: 21

 Total number of segments 124
 Number of rejected segments 22

 Segmentation in Complete.

 Feature extraction in process...
 Feature extraction in complete.

 Number of good grain : 84
 Number Not good grain or imputity: 18

About:

Please feel free to email & contact me if you run into issues or just would like to talk about the future usage.

wheat-quality-detector-2's People

Contributors

Stargazers

Watchers

Forkers

lavanya02011999 haitian-zhang itisdb

wheat-quality-detector-2's Issues

Grain (Paddy) Detect from Image

Detect Paddy in Image

Count the number of paddy in a image.
I have create a Model to detect paddy using Tensorflow API Object detection model but it did not work properly, can please help me out how to do that.
I have learn a lot from your example i.e. Wheat-quality-detector.
Please tell which approach are you using and technology and some tutorial and blog are you have pls suggest.

When Run python cmd_wheat_quality_detector_v2.py

It will give me error :

`Using TensorFlow backend.
2019-10-09 10:43:05.765390: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 3092785000 Hz
2019-10-09 10:43:05.765957: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x555efc684df0 executing computations on platform Host. Devices:
2019-10-09 10:43:05.765992: I tensorflow/compiler/xla/service/service.cc:175] StreamExecutor device (0): Host, Default Version
Enter the file(wheat image) location to dectect : mark.jpg
Traceback (most recent call last):
File "/media1/miniconda3/envs/test_env_36/lib/python3.6/site-packages/keras/engine/saving.py", line 382, in _deserialize_model
model.optimizer.set_weights(optimizer_weight_values)
File "/media1/miniconda3/envs/test_env_36/lib/python3.6/site-packages/keras/optimizers.py", line 126, in set_weights
'of the optimizer (' + str(len(params)) + ')')
ValueError: Length of the specified weight list (13) does not match the number of weights of the optimizer (19)

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "cmd_wheat_quality_detector_v2.py", line 20, in
model = keras.models.load_model('weights_results_5out/weights_01234567.pkl')
File "/media1/miniconda3/envs/test_env_36/lib/python3.6/site-packages/keras/engine/saving.py", line 492, in load_wrapper
return load_function(*args, **kwargs)
File "/media1/miniconda3/envs/test_env_36/lib/python3.6/site-packages/keras/engine/saving.py", line 584, in load_model
model = _deserialize_model(h5dict, custom_objects, compile)
File "/media1/miniconda3/envs/test_env_36/lib/python3.6/site-packages/keras/engine/saving.py", line 384, in _deserialize_model
warnings.warn('Error in loading the saved optimizer '
UserWarning: Error in loading the saved optimizer state. As a result, your model is starting with a freshly initialized optimizer.
`
Thank You

Detect single paddy from image :

How to count number of paddy in image: