def SIFTgrain(self, octave): path_analyse = '/CT_analysesClassification/' + octave octaves = 6 initial_sigma = 1.1 gauss = GaussianSmoothing3D(self.path + path_analyse, octaves).smoothing(initial_sigma) dimension = 3 dog = DoG(self.path + path_analyse, dimension) dog.apply() pathDoG = '/3DDoG/' local_extrema = LocalExterma3D(self.path + path_analyse + pathDoG, self.path + path_analyse, True) local_extrema.find() path_local_extrema = '/3DLocalExtremum/' extrema = ExtremaSpace3D(self.path + path_analyse + pathDoG + path_local_extrema) extrema.find() Hessian = HessianMatrix(5.0) Hessian.HessianElimination(self.path + path_analyse) keypointorientation = KeyPointOrientation(self.path + path_analyse) keypointorientation.apply() path_keypoint_orientation = '/KeyPointsOrientation/' images = ReadImage(self.path + path_analyse + path_keypoint_orientation).openImage() for im in images: rotateImage(im, 10, self.path + path_analyse).apply() path_discriptor = '/Descriptor3D/' keydis = KeypointsFeatures(self.path + path_analyse + path_discriptor, self.path + path_analyse).apply()
def SIFTgrain(self, octave): path_analyse = '/CT_analysesClassification/' + octave octaves = 6 initial_sigma = 1.1 gauss = GaussianSmoothing3D(self.path + path_analyse, octaves).smoothing(initial_sigma) dimension = 3 dog = DoG(self.path + path_analyse, dimension) dog.apply() pathDoG = '/3DDoG/' local_extrema = LocalExterma3D(self.path + path_analyse + pathDoG, self.path + path_analyse, True) local_extrema.find() path_local_extrema = '/3DLocalExtremum/' extrema = ExtremaSpace3D(self.path + path_analyse + pathDoG + path_local_extrema) extrema.find() Hessian = HessianMatrix(5.0) Hessian.HessianElimination(self.path + path_analyse) keypointorientation = KeyPointOrientation(self.path + path_analyse) keypointorientation.apply() path_keypoint_orientation = '/KeyPointsOrientation/' images = ReadImage(self.path + path_analyse + path_keypoint_orientation).openImage() for im in images: rotateImage(im, 10, self.path + path_analyse).apply() path_discriptor = '/Descriptor3D/' keydis = KeypointsFeatures(self.path + path_analyse + path_discriptor, self.path + path_analyse).apply()
def test_case3(): assert rotateImage.rotateImage([[10, 9, 6, 3, 7], [6, 10, 2, 9, 7], [7, 6, 3, 8, 2], [8, 9, 7, 9, 9], [6, 8, 6, 8, 2]]) == [[6, 8, 7, 6, 10], [8, 9, 6, 10, 9], [6, 7, 3, 2, 6], [8, 9, 8, 9, 3], [2, 9, 2, 7, 7]]
def findTestType2(imageName, imageInitialName, dimX, dimY): # Se citeste imaginea image = cv2.imread(imageInitialName) # Se determina cooronatele celor 8 colturi ale grilelor existente in imagine (x1, y1, x2, y2, x3, y3, x4, y4, x5, y5, x6, y6, x7, y7, x8, y8, _) = rotateImage(imageName) valueX = int((x8 * dimX) / 620) valueY = int((y8 * dimY) / 870) imageTestType = image[valueY - 280:valueY - 3, valueX - 120:valueX + 10] # Se decupeaza locul in care se gasesc cele doua casute pentru bifa de Fizica / Informatica, din imaginea initiala # care a fost citita in variabila image cv2.imwrite('images/imgType.jpg', imageTestType)
def findTestType(imgName): # Se defineste un filtu de sharpening al imaginii kernelSharpening = np.array([[-1, -1, -1], [-1, 9, -1], [-1, -1, -1]]) # Se citeste imaginea image = cv2.imread(imgName) # Se redimensioneaza imaginea image = cv2.resize(image, (620, 870), interpolation=cv2.INTER_AREA) # Se determina cooronatele celor 8 colturi ale grilelor existente in imagine (x1, y1, x2, y2, x3, y3, x4, y4, x5, y5, x6, y6, x7, y7, x8, y8, rotateNumber) = rotateImage(imgName) # Se decupeaza locul in care se gasesc cele doua casute pentru bifa de Fizica / Informatica imageTestType = image[y8 - 65:y8 - 3, x8 - 35:x8 + 10] # Se aplica imaginii filtrul de accentuare a detaliilor, kernelSharpening, definit anterior imageTestType = cv2.filter2D(imageTestType, -1, kernelSharpening) # Se prelucreaza acesta imagine decupata pentru calcularea maximului intensitatilor pixelilor # Se transforma imaginea din RGB in imagine in tonuri de gri grayImageTestType = cv2.cvtColor(imageTestType, cv2.COLOR_BGR2GRAY) # Se aplica un filtru de blurare a imaginii grayImageTestType = cv2.GaussianBlur(grayImageTestType, (3, 3), 0) # Dorim acum detectarea marginilor din imagine, folosind functia Canny, asupra imaginii in tonuri de gri edgedImageTestType = cv2.Canny(grayImageTestType, 10, 250) # Vom construi un nucleu 'kernel' folosind functia getStructuringElement care va returna un element de o # anumita dimensiune si forma pentru anumite operatii morfologice kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (3, 3)) # In continuare vom aplica nucleul creat la pasul anterior intr-o transformare morfologica de tipul specificat # prin parametrul MORPH_CLOSE (operatie de inchidere), asupra imaginii produse de aplicarea edge detectorului # Canny. Functia folosita pentru aceasta transformare este morphologyEx closedImageTestType = cv2.morphologyEx(edgedImageTestType, cv2.MORPH_CLOSE, kernel) lenY = closedImageTestType.shape[0] # Se calculeaza suma pixelilor din jumatatea de sus, respectiv jumatatea de jos in variabilele sumPixels1, # sumPixels2 sumPixels1 = 0 for i in range(0, int(lenY / 2)): sumPixels1 = sumPixels1 + np.sum(np.sum(closedImageTestType[i])) sumPixels2 = 0 for i in range(int(lenY / 2), lenY): sumPixels2 = sumPixels2 + np.sum(np.sum(closedImageTestType[i])) # Daca maximul este sumPixels1, adica in jumtatatea de sus => raspunsul este Informatica; # altfel => raspunsul este Fizica if sumPixels1 > sumPixels2: return "Informatica" else: return "Fizica"
rotationDuration = 5 ''' rotationDuration is the length of time the rotation occurs between certain key frames example image starts at 0 secs and goes until 1 sec and flips one time. rotationDuration would be 1 (for 1 sec) image starts at 0 secs and goes until 2 secs BUT there's a change in the roation at the 1 sec mark roationDuration would need to be set twice and would be 1 sec both times. ''' rotationDurationPerFrame = round(float(rotationDuration)/frameRate,6) amountToRatePerFrame = float(rotation)/frameRate startingRotation = 0 while (startTime < rotationDuration): rotatedImage = rotateImage.rotateImage(rotatingImage,startTime,rotationDurationPerFrame,startingRotation) countingRotation = countingRotation + 1 startTime = startTime + rotationDurationPerFrame startingRotation = startingRotation + amountToRatePerFrame imageArray.append(rotatedImage) final = CompositeVideoClip(imageArray, size = moviesize) final.set_duration(5).write_videofile("exports/rotationTest_5secs.mp4", fps=frameRate,codec='mpeg4',bitrate="4000k",audio_codec="mp3")
def test_case1(): assert rotateImage.rotateImage([[1, 2, 3], [4, 5, 6], [7, 8, 9]]) == [[7, 4, 1], [8, 5, 2], [9, 6, 3]]
def test_case2(): assert rotateImage.rotateImage([[1]]) == [[1]]
def setUp(self): self.open = ReadImage( './test_data/1_nd/CT_analyses/KeyPointsOrientation/').openImage() self.rotate = rotateImage(self.open[0], 10, './test_data/1_nd/CT_analyses/')
def setUp(self): self.open = ReadImage('./test_data/1_nd/CT_analyses/KeyPointsOrientation/').openImage() self.rotate = rotateImage(self.open[0],10,'./test_data/1_nd/CT_analyses/')