def test_custom_recognition_full_model_resnet_multi():
try:
keras.backend.clear_session()
except:
None
predictor = CustomImagePrediction()
predictor.setModelPath(os.path.join(main_folder, "data-models", "idenprof_full_resnet_ex-001_acc-0.119792.h5"))
predictor.setJsonPath(model_json=os.path.join(main_folder, "data-json", "idenprof.json"))
predictor.loadFullModel(num_objects=10)
images_to_image_array()
result_array = predictor.predictMultipleImages(sent_images_array=all_images_array)
assert isinstance(result_array, list)
for result in result_array:
assert "predictions" in result
assert "percentage_probabilities" in result
assert isinstance(result["predictions"], list)
assert isinstance(result["percentage_probabilities"], list)
assert isinstance(result["predictions"][0], str)
assert isinstance(result["percentage_probabilities"][0], str)
def test_custom_recognition_model_resnet_multi():
predictor = CustomImagePrediction()
predictor.setModelTypeAsResNet()
predictor.setModelPath(
os.path.join(main_folder, "data-models", "idenprof_resnet.h5"))
predictor.setJsonPath(
model_json=os.path.join(main_folder, "data-json", "idenprof.json"))
predictor.loadModel(num_objects=10)
images_to_image_array()
result_array = predictor.predictMultipleImages(
sent_images_array=all_images_array)
assert isinstance(result_array, list)
for result in result_array:
assert "predictions" in result
assert "percentage_probabilities" in result
assert isinstance(result["predictions"], list)
assert isinstance(result["percentage_probabilities"], list)
assert isinstance(result["predictions"][0], str)
assert isinstance(result["percentage_probabilities"][0], str)
text = pytesseract.image_to_string(value, config='')
st.write("The text detected is")
st.write(text)
st.markdown("-----------------------------------")
st.title("Results")
if runResults == True:
if predicitions is not 0:
prob = round(float(probabilities[0]))
if prob > threshold:
st.write("The image ", singlePicTest, " has a head unit",
"with a ", prob, "probability.")
os.chdir(videopath)
if run == "Run Entire Video":
results_array = prediction.predictMultipleImages(finalpicnames,
result_count_per_image=2)
dfyy = pd.DataFrame(results_array,
columns=['HeadUnit', 'Probability', 'Frame'])
save_results = []
count = -1
for i in results_array:
headunit, prob = i["predictions"], i["percentage_probabilities"]
for idx in range(len(headunit)):
# print(pred[idx] , " : " , prob[idx])
if (headunit[idx] == "HeadUnit"):
count += 1
save_results.append((headunit[idx], prob[idx], count))
df = pd.DataFrame(save_results,
columns=['HeadUnit', 'Probability', 'Frame'])
from imageai.Prediction.Custom import CustomImagePrediction
from os import listdir, getcwd
#Searches for trained models
models = listdir("data/models")
#If no model exists, raise an error
if len(models) < 2: #(2 because .gitignore always exists at the folder)
raise Exception("No trained model found")
models.sort(reverse=True)
bestModel = models[0]
prediction = CustomImagePrediction()
prediction.setModelTypeAsResNet()
prediction.setModelPath(f"data/models/{bestModel}")
prediction.setJsonPath("data/json/model_class.json")
prediction.loadModel(num_objects=2)
#Searches for files to predict
filenames = listdir("predictFiles")
filenames.remove(".gitignore")
filepaths = ["predictFiles/" + filename for filename in filenames]
results = prediction.predictMultipleImages(filepaths)
for filename, result in zip(filenames, results):
print(f"{filename}: {result['predictions'][0]} ({result['percentage_probabilities'][0]}% certainty)")
def detection(num):
bound = []
index = []
main_index = []
ls = []
img = []
res = []
item = ''
classes = []
count = 0
counter = []
execution_path = os.getcwd()
model_name = os.listdir('Dataset/models/')
model_name.sort()
a = model_name[len(model_name) - 1]
prediction = CustomImagePrediction()
prediction.setModelTypeAsResNet()
prediction.setModelPath(
os.path.join(execution_path, 'Dataset/models/' + str(a)))
prediction.setJsonPath(
os.path.join(execution_path, 'Dataset/json/model_class.json'))
prediction.loadModel(num_objects=num)
for i in os.listdir(execution_path + '/Dataset/train'):
classes.append(i)
count = count + 1
os.chdir(os.getcwd() + '/images/crop')
exec_path = os.getcwd()
counter = [0] * count
try:
duck = os.listdir('.')
duck.sort()
res = prediction.predictMultipleImages(duck, result_count_per_image=1)
for each_result in res:
predictions, percentage_probabilities = each_result[
"predictions"], each_result["percentage_probabilities"]
img.append(predictions)
for j in range(0, len(img)):
for k in range(0, count):
if "['" + str(classes[k]) + "']" == str(
img[j]) or '["' + str(classes[k]) + '"]' == str(
img[j]):
counter[k] = counter[k] + 1
for j in range(0, count):
for k in range(0, len(img)):
if "['" + str(classes[j]) + "']" == str(img[k]):
index.append(k)
main_index.append(index)
index = []
for i in range(0, count):
for j in range(0, counter[i] - 1):
if main_index[i][j] + 1 != main_index[i][j + 1]:
bound.append(str(classes[i]))
os.chdir(execution_path)
counter.insert(0, str(datetime.datetime.now())[:-10])
try:
if bound != []:
with open('Bound_Error.csv', 'w') as csv_file:
writer = csv.writer(csv_file, delimiter=',')
writer.writerow(bound)
else:
os.remove('Bound_Error.csv')
except:
print('No error in placement ')
with open("Results.csv", "a") as csv_file:
writer = csv.writer(csv_file, delimiter=',')
writer.writerow(counter)
except Exception as e:
print(e)
print('Unknown object detected')
os.path.join(execution_path, "model_ex-091_acc-0.925676.h5"))
prediction.setJsonPath(os.path.join(execution_path, "model_class.json"))
prediction.loadModel(num_objects=13)
# for single image
# predictions, probabilities = prediction.predictImage(os.path.join(execution_path, "test_image/22.jpg"), result_count=5)
#
# for eachPrediction, eachProbability in zip(predictions, probabilities):
# print(eachPrediction , " : " , eachProbability)
# end
# for multiple image
all_images_array = []
for filename in glob.glob('E:/Python Code/CarLogo/venv/test_image/*.jpg'):
im = Image.open(filename)
all_images_array.append(im)
i = 0
results_array = prediction.predictMultipleImages(all_images_array,
result_count_per_image=1,
input_type='array')
for each_result in results_array:
predictions, percentage_probabilities = each_result[
"predictions"], each_result["percentage_probabilities"]
for index in range(len(predictions)):
print(predictions[index], " : ", percentage_probabilities[index])
print("-----------------------")
plt.imshow(np.asarray(all_images_array[i]))
plt.show()
i = i + 1
